giulia pergolizzi - welcome to uea digital repository ... · method (pathway b) 79 2.7.2...
TRANSCRIPT
Novel derivatives of nicotinamide
adenine dinucleotide (NAD) and their
biological evaluation against NAD-
Consuming Enzymes
Giulia Pergolizzi
University of East Anglia
School of Pharmacy
Thesis submitted for the degree of Doctor of Philosophy
July, 2012
© This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that
its copyright rests with the author and that use of any information derived there from must be in accordance with
current UK Copyright Law. In addition, any quotation or extract must include full attribution.
i
ABSTRACT
Nicotinamide adenine dinucleotide (β-NAD+) is a primary metabolite involved in
fundamental biological processes. Its molecular structure with characteristic functional
groups, such as the quaternary nitrogen of the nicotinamide ring, and the two high-
energy pyrophosphate and nicotinamide N-glycosidic bonds, allows it to undergo
different reactions depending on the reactive moiety. Well known as a redox substrate
owing to the redox properties of the nicotinamide ring, β-NAD+ is also fundamental as a
substrate of NAD+-consuming enzymes that cleave either high-energy bonds to catalyse
their reactions. In this study, a panel of novel adenine-modified NAD+ derivatives was
synthesized and biologically evaluated against different NAD+-consuming enzymes.
The synthesis of NAD+ derivatives, modified in position 2, 6 or 8 of the adenine ring
with aryl/heteroaryl groups, was accomplished by Suzuki-Miyaura cross-couplings.
Their biological activity as inhibitors and/or non-natural substrates was assessed against
a selected range of NAD+-consuming enzymes. The fluorescence of 8-aryl/heteroaryl
NAD+ derivatives allowed their use as biochemical probes for the development of
continuous biochemical assays to monitor NAD+-consuming enzyme activities. The
introduction of different substituents in position 8 on the adenine ring allowed the
modulation of their fluorescence, resulting in the development of more sensitive and
alternative probes compared to the known fluorophore ε-NAD+. The different
substitutions introduced on the adenine ring also allowed us to probe the active site of
an NAD+-dependent bacterial DNA ligase. The selective activity of 8-aryl/heteroaryl
NAD+ derivatives against different NAD
+-consuming enzymes offers excellent
opportunities for their application as tool compounds in in-vitro/in-vivo studies, and as
inhibitor templates for drug discovery.
ii
Parts of this thesis have already been published in scientific journals, and presented at
scientific conferences.
A novel fluorescent probe for NAD-consuming enzymes
Giulia Pergolizzi, Julea N. Butt, Richard P. Bowater and Gerd K. Wagner
Chemical Communications, 2011, 47, 12655-12657.
Two-Step Synthesis of Novel, Bioactive Derivatives of the Ubiquitous Cofactor
Nicotinamide Adenine Dinucleotide (NAD)
Thomas Pesnot, Julia Kempter, Jörg Schemies, Giulia Pergolizzi, Urszula Uciechowska,
Tobias Rumpf, Wolfgang Sippl, Manfred Jung, and Gerd K. Wagner
Journal of Medicinal Chemistry, 2011, 54, 3492-3499.
3th
– 8th
July 2011: Synthesis and biological evaluation of novel fluorescent NAD+
analogues. Giulia Pergolizzi, oral communication, European School of Medicinal
Chemistry (ESMEC) (Urbino, Italy).
April 2011: Base-modified NAD+ derivatives – Novel Chemical Tools for the
Investigation of NAD+-consuming Enzymes. Giulia Pergolizzi, Julea N. Butt, Richard
P. Bowater and Gerd K. Wagner, poster presentation, RSC Carbohydrate and
Bioorganic Meeting (5th
April) & RSC Bioorganic Postgraduate Symposium (15th
April)
(London, UK).
July – September 2010: Novel NAD+ derivatives modified at the adenine base:
synthesis and biological applications. Giulia Pergolizzi, Julea N. Butt, Richard P.
Bowater and Gerd K. Wagner, poster presentation, 6th
Nucleic Acids Forum Meeting
(2nd
July, Liverpool, UK) & 8th
NACON (Nucleic Acid Conference) (12th
– 16th
September, Sheffield, UK).
iii
TABLE OF CONTENTS
List of Figures
List of Schemes
List of Tables
List of Figures and Tables in Appendix
List of Abbreviations
Acknowledgements
1 Β-NAD+: AN ESSENTIAL BIOLOGICAL MOLECULE 1
1.1 β-NAD+ structure 1
1.1.1 Redox properties of β-NAD+ 4
1.1.2 β-NAD+ catabolism and NAD+-consuming enzymes 6
1.1.3 β-NAD+ biosynthesis 10
1.2 NAD+-consuming enzymes 13
1.2.1 NAD+-consuming enzymes: similarities and relationships 14
1.2.2 Intracellular and extracellular β-NAD+ balance 20
1.3 NAD+ derivatives as mechanistic probes for NAD+-dependent enzymes 23
1.3.1 NAD+ derivatives as inhibitors of NAD+-dependent enzymes 26
1.3.1.1 Nicotinamide modification 26
1.3.1.2 Nicotinamide ribose modification 28
1.3.1.3 Pyrophosphate bond modification 29
1.3.1.4 Adenine modification 31
1.3.2 NAD+ derivatives as non-natural substrates of NAD+-dependent enzymes 32
1.4 Assays for NAD+-consuming enzymes 35
1.4.1 Cyanide and dehydrogenase assays 36
1.4.2 -NAD+ and its application for NAD+-consuming enzyme assays 37
1.4.3 NGD+, NHD+ and NXD+ for ADPRC fluorescence-based assays 40
iv
1.4.4 6 and 8-alkyne NAD+ for NADase activity assays 41
1.4.5 β-NAD+ quantification by reaction with acetophenone 43
1.4.6 A continuous colourimetric assay for the detection of PARP activity 43
1.4.7 Alternative assays for the detection of sirtuins activity 44
1.4.8 Assays for the detection of DNA ligase activity 45
1.5 Project Objectives 46
2 SYNTHESIS OF ARYL/HETEROARYL ADENINE-MODIFIED NAD+ DERIVATIVES 50
2.1 Synthetic strategies to aryl/heteroaryl adenine-modified NAD+ derivatives 50
2.2 Substitution of the adenine ring by palladium-mediated cross-couplings 52
2.3 Synthetic route to 2-aryl/heteroaryl AMP derivatives 56
2.3.1 Iodination of C-2 and 2′,3′-O-isopropylidene protection of adenosine 57
2.3.2 Suzuki-Miyaura and Stille cross-couplings on C-2 of 2′,3′-O-isopropylidene-2-iodo adenosine 59
2.3.3 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine and subsequent deprotection to 2-aryl/heteroaryl AMP derivatives 61
2.4 Synthetic route to 6-aryl/heteroaryl AMP derivatives 63
2.4.1 Suzuki-Miyaura cross-coupling on C-6 of 6-chloro purine nucleoside 64
2.4.2 2′,3′-O-isopropylidene protection and phosphorylation of 5′-OH of 6-substituted purine nucleosides, with subsequent deprotection to 6-aryl/heteroaryl AMP derivatives 66
2.5 Synthetic route to 8-aryl/heteroaryl AMP derivatives 67
2.5.1 Bromination of C-8 of adenosine and adenosine monophosphate 69
2.5.2 Suzuki-Miyaura cross-coupling on C-8 of 8-bromo adenosine monophosphate to 8-aryl/heteroaryl AMP derivatives 70
2.6 Synthesis of aryl/heteroaryl adenine-modified NAD+ derivatives (pathway a) 76
2.6.1 AMP activation via phosphoromorpholidate intermediate 76
2.6.2 Coupling with β-NMN 77
v
2.7 Alternative synthetic strategies to aryl/heteroaryl adenine-modified NAD+ derivatives 79
2.7.1 Aryl/heteroaryl adenine-modified NAD+ derivatives via phosphoramidite method (pathway b) 79
2.7.2 Aryl/heteroaryl adenine-modified NAD+ derivatives via one-pot, two-step procedure (pathway c) 80
2.8 Conclusions 82
3 CONFORMATIONAL AND SPECTROSCOPIC ANALYSIS OF ARYL/HETEROARYL ADENINE-MODIFIED AMP AND NAD+ DERIVATIVES 83
3.1 Conformational analysis of nucleosides and nucleotides 84
3.1.1 Conformation of aryl/heteroaryl adenine-modified AMP derivatives 86
3.1.2 Conformation of aryl/heteroaryl adenine-modified NAD+ derivatives 87
3.2 Nucleobases and their photochemical properties 88
3.3 Design of emissive nucleosides 90
3.3.1 Fluorophore features 90
3.3.2 Families of existing nucleobase-modified emissive nucleosides 91
3.4 Measurements of quantum yield for aryl/heteroaryl adenine-modified AMP and NAD+ derivatives 94
3.5 Spectroscopic analysis of aryl/heteroaryl adenine-modified AMP derivatives 95
3.6 Spectroscopic analysis of aryl/heteroaryl adenine-modified NAD+ derivatives 101
3.7 Fluorescence quenching in 8-aryl/heteroaryl NAD+ derivatives 106
3.7.1 Fluorescence basis for internal quenching in NAD+ derivatives 106
3.7.2 NMR basis for internal quenching in NAD+ derivatives 111
3.8 Conclusions 117
vi
4 EVALUATION OF 8-ARYL/HETEROARYL NAD+ DERIVATIVES AS FLUORESCENT PROBES FOR NAD+-CONSUMING ENZYMES 120
4.1 Nucleotide pyrophosphatases (NPPs) 121
4.1.1 Fluorescence-based assay for nucleotide pyrophosphatase 123
4.1.2 Kinetic analysis of NPP activity 129
4.2 NAD+-glycohydrolases (NADases) and ADP-ribosyl cyclases (ADPRCs) 131
4.2.1 Fluorescence-based assay for NAD+-glycohydrolase 133
4.2.2 Fluorescence-based assay for ADP-ribosyl cyclase 135
4.2.3 Mechanistic aspects of ADPRC activity 139
5 EVALUATION OF ARYL/HETEROARYL ADENINE-MODIFIED NAD+ DERIVATIVES AS INHIBITORS AND NON-NATURAL SUBSTRATES FOR NAD+-CONSUMING ENZYMES 145
5.1 NAD+-dependent DNA ligases 145
5.1.1 Expression, purification and activity of Escherichia coli DNA ligase 153
5.1.2 Biological testing of 2-substituted AMP and NAD+ derivatives 155
5.1.2.1 Biological activities of 2-substituted AMP and NAD+ derivatives 160
5.1.3 Biological testing of 6-substituted AMP and NAD+ derivatives 162
5.1.3.1 Biological activities of 6-substituted AMP and NAD+ derivatives 163
5.1.4 Biological testing of 8-substituted NAD+ derivatives 164
5.1.4.1 Biological activities of 8-substituted NAD+ derivatives 166
5.1.5 Summary 168
5.2 Sirtuins 170
5.2.1 Biological testing of 8-substituted NAD+ derivatives 172
5.2.2 Biological activities of 8-substituted NAD+ derivatives 174
6 CONCLUSIONS AND FUTURE WORK 176
vii
7 EXPERIMENTAL 184
7.1 General 184
7.1.1 Synthesis and characterization 184
7.1.2 Chromatography conditions 185
7.1.3 Enzymology 186
7.2 Synthesis 186
7.2.1 General synthetic procedures 186
7.2.2 Preparation of 2-substituted AMP derivatives (7a-c) 189
7.2.3 Preparation of 6-substituted AMP derivatives (8a-b) 198
7.2.4 Preparation of 8-substituted AMP derivatives (9b-n) 205
7.2.5 Preparation of aryl/heteroaryl adenine-modified NAD+ derivatives (13a-b, 14a-b, 15b,c,e,g,h) 215
7.2.6 Synthesis of 8-(pyrrol-2-yl) cADPR (24) 229
7.3 Quantum yield measurements for AMP and NAD+ derivatives 230
7.4 Chemical hydrolysis of NAD+ derivatives 230
7.5 Protocols for HPLC assays of enzyme activities 231
7.5.1 Nucleotide pyrophosphatase (NPP) 231
7.5.2 NAD+-glycohydrolase (NADase) 231
7.5.3 ADP-ribosyl cyclase (ADPRC) 232
7.6 Protocols for fluorescent assays of enzyme activities 232
7.6.1 Nucleotide pyrophosphatase (NPP) 232
7.6.2 NAD+-glycohydrolase (NADase) 234
7.6.3 ADP-ribosyl cyclase (ADPRC) 234
7.7 Escherichia coli DNA ligase (EcLigA) 235
7.7.1 Expression and purification 235
7.7.2 Preparation of DNA substrate 236
7.7.3 Ligation assay 237
viii
7.7.4 Gel electrophoresis 238
7.7.5 Molecular modeling for DNA ligase 238
7.8 Sequence alignment and 3D alignment of protein structures 239
8 APPENDIX 240
9 REFERENCES 248
ix
List of Figures
Figure 1.1 β-NAD+ structure. ............................................................................................ 1
Figure 1.2 β-NAD+/β-NADH redox reaction. ................................................................... 2
Figure 1.3 β-NAD+ and its high-energy bonds. ................................................................ 3
Figure 1.4 β-NAD+ decomposition pathways in basic conditions. ................................... 4
Figure 1.5 Glucose metabolism and enzymatic reactions involving β-NAD+(H) as a
redox substrate. ................................................................................................................. 5
Figure 1.6 NAD+ derivatives from NAD
+-consuming enzymes activities and their
functions. ........................................................................................................................... 8
Figure 1.7 Catabolic processes (blue arrow) mainly involving the reduction of β-NAD+
to β-NADH; anabolic processes (red arrow) involving the oxidation of β-NADPH to β-
NADP+. ........................................................................................................................... 10
Figure 1.8 β-NAD+ biosynthesis and metabolism. .......................................................... 11
Figure 1.9 Formation of an oxocarbenium intermediate during the NAD+-
glycohydrolase activity of NAD+-consuming enzymes. ................................................. 15
Figure 1.10 β-NAD+ metabolism and its influence on the activity of sirtuins. ............... 17
Figure 1.11 Inhibition/activation of sirtuins by the activity of PARPs in response to
DNA damage. .................................................................................................................. 19
Figure 1.12 Pathways for the extracellular/intracellular β-NAD+ balance. .................... 22
Figure 1.13 Relationship between group modifications on β-NAD+ structure and
specific enzyme responses. ............................................................................................. 25
Figure 1.14 Inhibitors of DHs, NADKs, ARTs, PARPs, sirtuins, ADPRCs and NADases
based on nicotinamide modifications. ............................................................................. 27
Figure 1.15 Inhibitors of DHs, NADKs, sirtuins and NADases based on nicotinamide
ribose modifications. ....................................................................................................... 29
Figure 1.16 Inhibitors of DHs, NADKs and NPPs based on pyrophosphate bond
modifications. .................................................................................................................. 30
Figure 1.17 Inhibitors of ADPRCs, NPPs, sirtuins and DNA ligases based on adenine
modifications. .................................................................................................................. 32
Figure 1.18 NAD+ derivatives with NNS activities toward NAD
+-dependent enzymes.
......................................................................................................................................... 33
Figure 1.19 Chemical structure of ε-NAD+. ................................................................... 34
x
Figure 1.20 Cyanide and dehydrogenase (DH) assays to detect NAD+-consuming
activities. ......................................................................................................................... 37
Figure 1.21 ε-NAD+ as a NNS of NADases, NPPs, ADPRCs and ARTs for the
development of continuous fluorescence-based assays. ................................................. 38
Figure 1.22 ε-NAD+ and its derivatives. ......................................................................... 39
Figure 1.23 NGD+, NHD
+ and NXD
+ as NNSs of ADPRCs for the development of
continuous fluorescence-based assays. ........................................................................... 41
Figure 1.24 6 and 8-alkyne NAD+ as NNSs of NADases, ARTs, PARPs and sirtuins for
the development of fluorescence or immunohistochemical assays. ................................ 42
Figure 1.25 Quantification of the unreacted β-NAD+ by its reaction with acetophenone
to generate the fluorescent NAD+ derivative, 1. ............................................................. 43
Figure 1.26 ADPR-para-nitrophenol (ADPR-pNP) as a NNS of PARPs for the
development of a continuous colourimetric assay. ......................................................... 44
Figure 1.27 Fluorescence-based assay for sirtuins activity. ............................................ 45
Figure 1.28 Ligation assays based on the generation of FRET signals (a), or on the
fluorescence detection of ligated molecular beacons105
(b) upon ligation and
denaturation conditions. .................................................................................................. 46
Figure 1.29 Synthetic (a) and biological (b) targets of the project. ................................ 48
Figure 2.1 Purine structure and numbering. .................................................................... 53
Figure 2.2 Ionization equilibrium of the 3-pyridinyl boronic acid in water. .................. 72
Figure 2.3 Acid-base titration of 3-(aminomethyl)phenyl AMP (9d) using a pH meter. ...
......................................................................................................................................... 74
Figure 2.4 1H NMR spectra of the two isolated conformers, a and b, for 9d. ................ 75
Figure 3.1 Glycosidic conformational ranges of nucleosides and nucleotides. .............. 84
Figure 3.2 Adenine conformations in the natural adenosine/AMP and the aryl/heteroaryl
adenine-modified adenosine/AMP derivatives. .............................................................. 85
Figure 3.3 Ribose conformations of nucleosides (R = H) and nucleotides (R = Pi). ...... 85
Figure 3.4 Physical processes occurring when a molecule absorbs UV/Vis light. ......... 89
Figure 3.5 Emissive nucleosides obtained by replacement of the nucleobases or
introduction on their skeleton of known fluorophores. ................................................... 92
Figure 3.6 Expanded and extended nucleosides, obtained by extension of the π-electrons
delocalization of the nucleobases. ................................................................................... 92
Figure 3.7 Emissive isomorphic nucleosides. ................................................................. 93
xi
Figure 3.8 Absorbance spectra of AMP and 9b,e,h in water. ......................................... 96
Figure 3.9 Variations of λex with the different substituents in water and methanol for the
8-aryl/heteroaryl adenine-modified AMP derivatives synthesized in this study. ........... 97
Figure 3.10 Fluorescence emission spectra of 9b,e,h in water at λex maxima (see Table
3.3). ................................................................................................................................. 97
Figure 3.11 Variations of λem, Stokes shift and quantum yield () values with the
different substituents in water and methanol for the 8-aryl/heteroaryl adenine-modified
AMP derivatives synthesized in this study. .................................................................... 98
Figure 3.12 Comparison of fluorescence of adenine-modified AMP derivatives with the
same/similar aryl/heteroaryl substituent in different positions. .................................... 100
Figure 3.13 Absorbance spectrum of 15e in water. ...................................................... 102
Figure 3.14 Fluorescence emission spectra for 15b,c,e,g,h (0.8 μM in water) at λex
maxima (see Table 3.5). ................................................................................................ 103
Figure 3.15 Protonation of N-1 of 8-aryl/heteroaryl adenine-modified NAD+
derivatives. .................................................................................................................... 103
Figure 3.16 Fluorescence emission at different pH for 15b,c,e,g,h. ............................. 104
Figure 3.17 Network of H-bonds around 15e and 15g, responsible for the red shift of λem
in 15e and 15g with the increase of the pH, and the hypochromic effect at basic pH in
15g. ................................................................................................................................ 106
Figure 3.18 ε-NAD+ and its fluorescence internal quenching. ..................................... 108
Figure 3.19 MeOH titration of 15c,e,g,h. ..................................................................... 110
Figure 3.20 Fluorescence time-course for the basic hydrolysis of NAD+ derivatives
15b,c,e,g,h. .................................................................................................................... 111
Figure 3.21 1H NMR analysis of aromatic protons in β-NAD
+ with increasing
percentage of ACN. ....................................................................................................... 113
Figure 3.22 1H NMR analysis of 1′-H and 1′′-H in β-NAD
+ with increasing percentage
of ACN. ......................................................................................................................... 114
Figure 3.23 1H NMR peaks of 1′-H and 1′′-H in β-NAD
+ with increasing percentages of
ACN. ............................................................................................................................. 114
Figure 3.24 1H NMR analysis of aromatic protons in 15h with increasing percentage of
ACN. ............................................................................................................................. 115
Figure 3.25 β-NAD+ and 15h folded conformations. ................................................... 116
Figure 3.26 1H NMR analysis of 1′-H and 1′′-H in 15h with increasing percentage of
ACN. ............................................................................................................................. 116
xii
Figure 3.27 Structures for -adenine and 8-aryl/heteroaryl adenine derivatives. ......... 119
Figure 4.1 HPLC time-course of the enzymatic reaction of NPP from Crotalus
adamanteus venom and 15h. ......................................................................................... 123
Figure 4.2 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h. .... 125
Figure 4.3 Michaelis-Menten plot of NPP from Crotalus adamanteus venom with 15h
as a substrate. ................................................................................................................ 126
Figure 4.4 IC50 estimation for β-NAD+ toward the activity of NPP from Crotalus
adamanteus venom and 15h. ........................................................................................ 127
Figure 4.5 Dixon plot for β-NAD+ inhibition toward the activity of NPP from Crotalus
adamanteus venom and 15h. ........................................................................................ 127
Figure 4.6 Dixon plot for AMP inhibition toward the activity of NPP from Crotalus
adamanteus venom and 15h. ........................................................................................ 128
Figure 4.7 HPLC time-course of the enzymatic reaction of NADase from porcine brain
and 15h. ......................................................................................................................... 134
Figure 4.8 Fluorimetric assay of NADase from porcine brain and 15h. ....................... 135
Figure 4.9 HPLC time-course of the enzymatic reaction of ADPRC from Aplysia
californica and 15h. ...................................................................................................... 136
Figure 4.10 Fluorescence changes observed during the enzymatic reaction between
ADPRC from Aplysia californica and 15h. .................................................................. 137
Figure 4.11 Fluorimetric assay of ADPRC from Aplysia californica and 15h: initial
velocity (V0) as function of ADPRC concentration. ..................................................... 138
Figure 4.12 Reaction mechanism for N-1 cyclization of 15h with ADPRC from Aplysia
californica. .................................................................................................................... 140
Figure 4.13 Cyclic ADP-ribose derivatives obtained by cyclization on N-1 or N-7 with
ADPRC from Aplysia californica. ................................................................................ 141
Figure 5.1 Structure of DNA ligase from E. coli (EcLigA) (PDB 2OWO) in complex
with nicked DNA. ......................................................................................................... 149
Figure 5.2 Hydrophobic tunnel on NTase domain of DNA ligase from Enterococcus
faecalis (PDB 1TAE). ................................................................................................... 151
Figure 5.3 C-2 adenine-modified adenosine derivatives and their efficacies as NAD+-
dependent DNA ligases inhibitors.73, 74, 263
................................................................... 152
Figure 5.4 Purification of EcLigA. ............................................................................... 153
Figure 5.5 Ligation assay with fluorescent annealed nicked substrate. ........................ 154
xiii
Figure 5.6 Denaturing PAGE for the NNS activity of 2-substituted AMP and NAD+
derivatives at 250 μM. ................................................................................................... 156
Figure 5.7 NNS activity of 2-substituted NAD+ derivatives at concentrations between 50
and 250 μM. .................................................................................................................. 156
Figure 5.8 Inhibition of 2-substituted AMP/AMP-morpholidate/NAD+ derivatives at
200 μM. ......................................................................................................................... 157
Figure 5.9 Competition of 2-substituted NAD+ derivatives, 13a-b (light blue), with β-
NAD+ (green) for EcLigA (E, brown). ......................................................................... 159
Figure 5.10 β-NAD+ (a) and 13b (b) in the active site of DNA ligase from E. faecalis
(PDB 1TAE). ................................................................................................................ 161
Figure 5.11 Inhibition of 6-substituted AMP/AMP-morpholidate/NAD+ derivatives at
250 μM. ......................................................................................................................... 163
Figure 5.12 NNS activity test of 15h. ........................................................................... 165
Figure 5.13 Inhibition of 8-substituted NAD+ derivatives at 250 μM. ......................... 165
Figure 5.14 β-NAD+ (a) and 15h (b) in the active site of DNA ligase from E. faecalis
(PDB 1TAE). ................................................................................................................ 166
Figure 5.15 Docking of 15h (pink) and superimposition with β-NAD+ (green) in the
active site of DNA ligase from E. faecalis (PDB 1TAE). ............................................ 167
Figure 5.16 a) Structure of archaeal Sir2-Af2 with β-NAD+ bound; b) β-NAD
+-binding
pocket (PDB 1YC2). ..................................................................................................... 172
Figure 5.17 8-aryl/heteroaryl NAD+ derivatives, 15b, 15h, 16-22, tested as inhibitors
and/or NNSs of human SIRT1-2.70
............................................................................... 173
Figure 6.1 Two possible fluorescent NNSs of NAD+-dependent DNA ligases:
nicotinamide 2-amino purine dinucleotide (1) and nicotinamide 2,6-diamino purine
dinucleotide (2). ............................................................................................................ 178
Figure 6.2 β-NAD+ in the active site of ADPRC from Aplysia californica (PDB 3ZWM)
(a), archaeal Sir2-Af2 (PDB 1YC2) (b), DNA ligase from E. faecalis (PDB 1TAE) (c).
....................................................................................................................................... 180
xiv
List of Schemes
Scheme 1.1 De novo pathway. ........................................................................................ 13
Scheme 2.1 General synthetic pathways to aryl/heteroaryl adenine-modified NAD+
derivatives. ...................................................................................................................... 51
Scheme 2.2 General synthetic strategy to aryl/heteroaryl adenine-modified nucleotides.
......................................................................................................................................... 54
Scheme 2.3 Catalytic cycle of palladium-mediated cross-couplings. ............................. 55
Scheme 2.4 Synthetic route to 2-substituted AMP derivatives. ...................................... 57
Scheme 2.5 Iodination of adenosine on C-2 from 1........................................................ 58
Scheme 2.6 2′,3′-O-isopropylidene protection of 2-iodo adenosine, 4. .......................... 59
Scheme 2.7 Palladium-mediated cross-couplings on C-2 of 2′,3′-O-isopropylidene-2-
iodo adenosine, 4a. .......................................................................................................... 61
Scheme 2.8 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine
by POCl3 (method 1). ...................................................................................................... 62
Scheme 2.9 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine
via phosphoramidite (method 2). .................................................................................... 63
Scheme 2.10 Synthetic route to 6-aryl/heteroaryl AMP derivatives. ............................. 64
Scheme 2.11 Suzuki-Miyaura cross-coupling on C-6 of 6-chloro purine nucleoside, 5. ...
......................................................................................................................................... 65
Scheme 2.12 Phosphorylation of 5′-OH and following deprotection of 2′,3′-O-
isopropylidene-6-substituted purine nucleosides. ........................................................... 66
Scheme 2.13 Synthetic route to 8-aryl/heteroaryl AMP derivatives. ............................. 68
Scheme 2.14 Bromination on C-8 of adenosine, 3, and AMP, 3a. ................................. 69
Scheme 2.15 Suzuki-Miyaura cross-coupling on C-8 of 8-bromo AMP, 9a. ................. 71
Scheme 2.16 Synthesis of 9d. ......................................................................................... 73
Scheme 2.17 Synthesis of aryl/heteroaryl adenine-modified AMP-morpholidate
derivatives. ...................................................................................................................... 76
Scheme 2.18 Synthesis of aryl/heteroaryl adenine-modified NAD+ derivatives. ........... 78
Scheme 2.19 Synthesis of 13b via phosphoramidite method (pathway b). .................... 80
Scheme 2.20 Synthesis of 8-aryl/heteroaryl adenine-modified NAD+ derivatives via
one-pot, two-step procedure (pathway c). ....................................................................... 81
Scheme 4.1 Catalytic mechanism of ENPPs. ................................................................ 122
xv
Scheme 4.2 Partitioning mechanism for Aplysia ADPRC, CD38 and NADases. ........ 133
Scheme 4.3 Synthetic pathway to 8-(pyrrol-2-yl) cADPR, 24. .................................... 143
Scheme 5.1 Reaction mechanism for DNA ligation by ATP and NAD+-dependent DNA
ligases. ........................................................................................................................... 147
List of Tables
Table 1.1 Intracellular metabolite concentrations in E. coli.20
.......................................... 9
Table 2.1 Synthesis of the 8-aryl/heteroaryl AMP derivatives (9b-n)............................ 70
Table 2.2 Synthesis of the aryl/heteroaryl adenine-modified NAD+ derivatives 13a-b,
14a-b, 15b,c,e,g,h. .......................................................................................................... 78
Table 3.1 Conformational analysis of the various aryl/heteroaryl adenine-modified
AMP derivatives synthesized in this study. .................................................................... 86
Table 3.2 Conformational analysis of the various aryl/heteroaryl adenine-modified
NAD+ derivatives synthesized in this study. ................................................................... 88
Table 3.3 Spectroscopic analysis in water and methanol of natural AMP and aryl-
heteroaryl adenine-modified AMP derivatives synthesized in this study. ...................... 95
Table 3.4 Comparative spectroscopic analysis of adenine-modified AMP derivatives
with the same substituent (phenyl and pyrrol-2-yl) in different positions. ................... 100
Table 3.5 Spectroscopic analysis in water and methanol of natural β-NAD+ and aryl-
heteroaryl adenine-modified NAD+ derivatives synthesized in this study. .................. 101
Table 3.6 pKa values for the natural β-NAD+ and 15b,c,e,h. ....................................... 105
Table 3.7 λem maxima of 15b,c,e,g,h at different pH. ................................................... 105
Table 3.8 Spectroscopic properties of ε-NAD+ and fluorescent amino acids. .............. 107
Table 3.9 Comparison of values in water and methanol for ε-AMP/ε-NAD+, and
aryl/heteroaryl adenine-modified AMP/NAD+ derivatives synthesized in this study. . 109
Table 3.10 λem maxima of 15b,d,f,g at different MeOH percentage. ........................... 110
Table 3.11 Comparison between ε-adenine and 8-aryl/heteroaryl adenine derivatives.
....................................................................................................................................... 118
Table 4.1 Enzymological properties for β-NAD+
, ε-NAD+ and the related dinucleotide
ε-PdAD+, and 15h with different nucleotide pyrophosphatases. .................................. 129
Table 4.2 Competitive inhibition of AMP and β-NMN toward different nucleotide
pyrophosphatases. ......................................................................................................... 130
xvi
Table 4.3 Enzymological properties for β-NAD+, 15h, ε-NAD
+ and other dinucleotide
substrates of ADPRC from Aplysia californica. ........................................................... 141
Table 4.4 Spectroscopic analysis and values of 15h and 24 in water. ...................... 144
Table 4.5 1H NMR analysis for 15h and 24. Comparison of chemical shift (ppm) for 2-
H, 1′-H and 1′′-H. .......................................................................................................... 144
Table 5.1 Essential residues in the NTase domain of EcLigA and their roles. ............. 150
Table 5.2 IC50 values ± SD for 7a and 13a toward EcLigA. ........................................ 157
Table 6.1 Activities of 8-aryl/heteroaryl NAD+ derivatives toward different NAD
+-
consuming enzymes. ..................................................................................................... 180
Table 6.2 Changes in the adenine conformation during the catalytic cycle in ADPRC,
SIRT1-2 and DNA ligase. ............................................................................................. 182
Table 7.1 Gradient steps for HPLC analysis of the NAD+ derivatives. ........................ 185
Table 7.2 Composition of buffers used for nickel-affinity chromatography. ............... 236
Table 7.3 Oligonucleotides sequence used for the preparation of substrate for ligation
assay. ............................................................................................................................. 236
List of Figures in Appendix
Figure A1 HPLC time-course for the basic hydrolysis of 15h. .................................... 240
Figure A2 a) Fluorescence calibration curves for 15h and 9h. b) Fluorimetric assay of
NPP from Crotalus adamanteus venom and 15h: initial velocity (V0) as function of
NPP concentration. ........................................................................................................ 240
Figure A3 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h: a)
Hanes-Woolf and b) Lineweaver-Burk plot. ................................................................ 241
Figure A4 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h:
Michaelis-Menten plot for β-NAD+ inhibition toward the activity of NPP from Crotalus
adamanteus venom and 15h. ........................................................................................ 241
Figure A5 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15b: a)
fluorescence time-course as fluorescence emission [RFU] vs time [sec]; b) Michaelis-
Menten plot. .................................................................................................................. 242
Figure A6 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15c:
fluorescence time-course as fluorescence emission [RFU] vs time [sec]. .................... 242
Figure A7 Fluorimetric assay of ADPRC from Aplysia californica and 15h: ADPR-
cyclase activity; fluorescence time-course as fluorescence emission [RFU] vs time [sec].
....................................................................................................................................... 243
xvii
Figure A8 Fluorimetric assay of ADPRC from Aplysia californica and 15h: Michaelis-
Menten and Hanes-Woolf plot (insert). ........................................................................ 243
Figure A9 Fluorimetric assay of ADPRC from Aplysia californica and 15h: cADPR-
hydrolase activity; fluorescence time-course as fluorescence emission [RFU] vs time
[sec]. .............................................................................................................................. 244
Figure A10 Comparison of hydrolytic stability for 15h and 24.................................... 244
Figure A11 Preparation of annealed substrate for ligation assay at different ratio
LigSub1:LigSub2:LigSub1+2. ...................................................................................... 245
Figure A12 IC50 estimation for quinacrine. .................................................................. 245
Figure A13 Denaturing PAGE for the inhibition of 2-substituted AMP/AMP-
morpholidate/NAD+ derivatives at 200 μM. ................................................................. 246
Figure A14 IC50 estimation for 7a (a) and 13a (b). ...................................................... 246
Figure A15 IC50 estimation for 11a. ............................................................................. 247
List of Tables in Appendix
Table A1 Structure pairwise alignment of ADPRC from Aplysia californica (PDB
3ZWM), archaeal Sir2-Af2 (PDB 1YC2), and DNA ligase from E. faecalis (PDB
1TAE) by rigid FATCAT and flexible FATCAT. ........................................................ 247
xviii
Abbreviations
quantum yield
chemical shift
°C degree Celsius
2′′(3′′)-OAc-ADPR 2′′(3′′)-acetate-adenosine diphosphate ribose
2′′-FNAD+ 2′′-fluoro-nicotinamide adenine dinucleotide (ox.form)
2-AP 2-amino purine
2-F-ADPR 2-fluoro adenosine diphosphate ribose
2-F-NAD+ 2-fluoro nicotinamide adenine dinucleotide (ox. form)
2HPC 2-hydroxy-3-pyridinecarboxaldehyde
3AcPAD+ 3-acetyl pyridine adenine dinucleotide (ox. form)
3AlPAD+ 3-aldehyde pyridine adenine dinucleotide (ox. form)
3HAO 3-hydroxyanthranilate-3,4-dioxygenase
Å angstrom
AAD bis-adenine dinucleotide
Ac acetyl
acetyl-CoA acetyl coenzyme A
ACN acetonitrile
AcOH acetic acid
Ado adenosine
ADP adenosine diphosphate
ADPR adenosine diphosphate ribose
ADPRC adenosine diphosphate ribosyl cyclase
ADPRNH2 1′′-amino adenosine diphosphate
ADPR-pNP adenosine diphosphate ribose para-nitro phenol
AMC 7-amino-4-methyl coumarin
AMP adenosine 5′-monophosphate
ara-2′′-FNAD+ nicotinamide 2′′-deoxy-2′′-fluoroarabinoside adenine
dinucleotide (ox. form)
ara-2′′-FNMN nicotinamide 2′′-deoxy-2′′-fluoroarabinoside mononucleotide
ara-NAD+ nicotinamide arabinoside adenine dinucleotide (ox. form)
Arg arginine
xix
ART adenosine diphosphate ribosyl transferase
ATP adenosine triphosphate
BAD benzamide adenine dinucleotide
Boc tert-butyloxycarbonyl
bp base pair
br s broad singlet
BRCT breast cancer carboxy-terminal
BSA bovin serum albumin
Bu butyl
cADPR cyclic adenosine diphosphate ribose
cADPRP 2′-phosphate cyclic adenosine diphosphate ribose
cAMP 3′,5′-cyclic adenosine monophosphate
carba-NAD+ carba-nicotinamide adenine dinucleotide (ox. form)
Cbz carboxybenzyl
cGDPR cyclic guanine diphosphate ribose
CICR Ca2+
-induced Ca2+
-release
cIDPR cyclic inosine diphosphate ribose
cm centimetre
CN cyanide
C-NAD C-nicotinamide adenine dinucleotide
C-PAD C-picolinamide adenine dinucleotide
cXDPR cyclic xanthosine diphosphate ribose
Cys cysteine
cεADPR cyclic 1,N6-etheno-adenosine diphosphate ribose
d doublet
dAMP deoxy adenosine monophosphate
DCI 4,5-dicyanoimidazole
dd doublet of doublets
DDR DNA damage response
dGMP deoxy guanosine monophosphate
DH dehydrogenase
DIPEA diisopropylethylamine
xx
DMAP 4-dimethylamino pyridine
DME dimethyl ether
DMF dimethylformamide
DMSO dimethylsulfoxide
DMT dimethoxy
DNA Lig deoxy ribonucleic acid ligase
DNA deoxy ribonucleic acid
DTT dithiothreitol
E glutamic acid
E. coli Escherichia coli
E. faecalis Enterococcus faecalis
EcLigA Escherichia coli DNA ligase
EDTA ethylenediaminetetraacetic acid
ENPP ecto-nucleotide pyrophosphatase
equiv equivalent
ESI electrospray ionization
EtOAc ethyl acetate
FAM 6-carboxy fluorescein
FRET Förster (fluorescence) resonance energy transfer
g gram
Grad gradient
HAT histone acetyl transferase
HCV hepatitis C virus
HDAC histone deacetylase
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
HHD bis-hypoxanthine dinucleotide
HhH helix-hairpin-helix
His histidine
HOBt hydroxybenzotriazole
HOMO highest occupied molecular orbital
HPLC high pressure liquid chromatography
HR-MS high resolution mass spectra
xxi
Hz hertz
IC internal conversion
IC50 half maximal inhibitory concentration
IDA iminodiacetic acid
IDO indoleamine-2,3-dioxygenase
Ile isoleucine
IP3 inositol 1,4,5-triphosphate
iPA isopropanol
IPTG isopropil β-D-1-tiogalattopiranoside
ISC inter system crossing
J coupling constant
K kelvin
K3H kynurenine 3-hydroxylase
kcal kilocalorie
kcat enzymatic turnover number
kDa kilodalton
KFase kynurenine formamidase
Ki inhibition constant for competitive inhibitor
Ki′ inhibition constant for uncompetitive inhibitor
Km Michaelis-Menten constant
Kyase kynureninase
LB Luria Bertani
LC liquid chromatography
LUMO lowest unoccupied molecular orbital
Lys lysine
M molar
m multiplet
m/z mass to charge ratio
MAL tert-butyloxycarbonyl 1-(4-methyl-2-oxo-2H-chromen-7-
ylcarbamoyl)-5-acetamidopentylcarbamate (Boc-Lys[Ac]-AMC)
Me methyl
MeOH methanol
xxii
mer polymer unit
mg milligram
MHz megahertz
mL millilitre
mM millimolar
mm millimetre
mol mole
MS mass spectra
N normal
NA nicotinic acid
NADase nicotinamide adenine dinucleotide glycohydrolase
NADK nicotinamide adenine dinucleotide kinase
NaM nicotinamide
NAMN nicotinic acid mononucleotide
NAMNAT nicotinic acid mononucleotide adenylyl transferase
NaMPRT nicotinamide phospho-ribosyl transferase
NaOAc sodium acetate
NAPRT nicotinic acid phospho-ribosyl transferase
NFD+ nicotinamide formycin dinucleotide (ox. form)
NGD+ nicotinamide guanine dinucleotide (ox. form)
NHD+ nicotinamide hypoxanthine dinucleotide (ox. form)
nm nanometre
nM nanomolar
NMNAT nicotinamide mononucleotide adenylyl transferase
NMR nuclear magnetic resonance
NNS non-natural substrate
NOE nuclear Overhauser effect
NOESY nuclear Overhauser effect spectroscopy
NPP nucleotide pyrophosphatase
NR nicotinamide ribonucleoside
NRK nicotinamide ribonucleoside kinase
NS natural substrate
xxiii
NTase nucleotidyl transferase
NXD+ nicotinamide xanthine dinucleotide (ox. form)
OAc acetate
OB oligonucleotide binding
OD optical density
PAGE polyacrylamide gel electrophoresis
PAH polycyclic aromatic hydrocarbon
PARG polyadenosine diphosphate ribosyl glycohydrolase
PARP polyadenosine diphosphate ribosyl polymerase
PDB protein data bank
Ph phenyl
Pi phosphate
pKa acid dissociation constant
ppb parts per billion
PPh3 triphenylphosphine
PPi diphosphate, or pyrophosphate
ppm parts per million
PPPi triphosphate
PTS polyoxyethanyl α-tocopheryl sebacate
q quartet
QA quinolic acid
QAPRT quinolic acid phospho-ribosyl transferase
RCSB research collaboratory for structural bioinformatics
Rf retention factor
RFU relative fluorescence unit
RNA ribonucleic acid
ROS reactive oxygen species
RP reverse phase
rt room temperature
s singlet
SD standard deviation
SDS sodium dodecyl sulphate
xxiv
SEAr aromatic electrophilic substitution
Sir silent information regulator
SIRT sirtuin
SN1(2) nucleophilic substitution
SNAr aromatic nucleophilic substitution
T threonine
t1/2 half life
TAMRA tetramethyl rhodamine
TBAF tetrabutylammonium fluoride
TBE tris/borate/EDTA
tBu tert-butyl
tBuOH tert-butanol
tBuOOH tert-butyl hydroperoxide
TDO tryptophan-2,3-dioxygenase
TEA triethylamine
TEAB triethylammonium bicarbonate buffer
TFA trifluoroacetic acid
TGS tris/glycine/SDS
THF tetrahydrofurane
thio-NAD+ thio-nicotinamide adenine dinucleotide (ox. form)
TIPS triisopropylsilyl
TLC thin layer chromatography
TPPTS triphenylphosphine trisulfonate
Tris tris(hydroxymethyl)aminomethane
tRNA transfer ribonucleic acid
Trp tryptophan
TXPTS trixylylphosphine trisulfonate
U enzyme unit
V volt
V0 initial velocity
Vmax maximum velocity
Vw van der Waals volume
xxv
W tryptophan
Y tyrosine
Z-MAL benzyl 1-(4-methyl-2-oxo-2H-chromen-7-ylcarbamoyl)-5-
acetamidopentylcarbamate (Cbz-Lys[Ac]-AMC)
ZML benzyl 1-(4-methyl-2-oxo-2H-chromen-7-ylcarbamoyl)-5-
aminopentylcarbamate (Cbz-Lys-AMC)
α-NAD+ α-nicotinamide adenine dinucleotide (ox. form)
β-NAAD+ β-nicotinic acid adenine dinucleotide (ox. form)
β-NAADP+ 2′-phosphate β-nicotinic acid adenine dinucleotide (ox. form)
β-NAD+ β-nicotinamide adenine dinucleotide (ox. form)
β-NADH β- nicotinamide adenine dinucleotide (red. form)
β-NADP+ 2′-phosphate β-nicotinamide adenine dinucleotide (ox. form)
β-NADPH 2′-phosphate β-nicotinamide adenine dinucleotide (red. form)
β-NMN β-nicotinamide mononucleotide
ΔE0 difference of standard reduction potential
ε-ADP 1,N6-etheno-adenosine diphosphate
ε-ADPR 1,N6-etheno-adenosine diphosphate ribose
ε-AMP 1,N6-etheno-adenosine monophosphate
ε-ATP 1,N6-etheno-adenosine triphosphate
ε-hy4PdAD
+ 1,N
6-etheno-4-hydrazinocarbonyl pyridine adenine dinucleotide
(ox. form)
ε-NAD+ 1,N
6-etheno-nicotinamide adenine dinucleotide
ε-PdAD+ 1,N
6-etheno-pyridine adenine dinucleotide (ox. form)
η refractive index
λem emission wavelength
λex excitation wavelength
μg microgram
μL microlitre
μm micrometre
μM micromolar
μmol micromole
ψ carba-NAD+
pseudo carba-nicotinamide adenine dinucleotide (ox. form)
xxvi
Acknowledgements
I would like to thank my supervisors, Dr Gerd Wagner, Dr Richard Bowater and Prof
Julea Butt, for encouraging, supporting and believing in me during the whole period of
my PhD, and for contributing to my knowledge as a scientist. I would also like to thank
UEA for the studentship and KCL for the financial support, which enabled me to carry
out this PhD.
I wish to thank all the teaching and technical staff at the KCL for their warm welcome
in the new institute. I wish to extend my gratitude to Andrew Atkinson (KCL), Shirley
Fairhurst (JIC), Colin MacDonald (UEA) and Martin Rejzek (JIC) for their assistance
with obtaining NMR data. I am grateful to the EPSCR National Mass Spectrometry
Service Centre in Swansea for the recording of Mass Spectra. I am also grateful to
Manfred Jung and his group for the biological testing of candidate molecules. I would
like to thank Sebastian Gehrke for its help in the docking studies, and Sarah Zähringer
for its help in the synthetic part of this project. Special thanks to Lauren for
proofreading this thesis.
I wish to thank my very special housemates and best friends, Doroty and Lorina, who
shared with me so many moments and emotions along these four years. Special thanks
go to Thomas for its great help as friend and mate in the laboratory, and to Simone for
being my “punch ball”.
I wish to thank all past and present members of the Wagner group: Alice, Andy E.,
Ferdinand, George, Karine, Lauren, Niina, Sebastian and Sarah for their help and
company in and outside the laboratory. And more, my colleagues, Andrew and Andy G.
for their support during my work in BIO (UEA). I would like to thank my dear friends,
Francesca and Andrea, who always made me smile, and all friends I have met along
these years, Alberto, Annalisa, Chiara, Eva, Ewelina, Fatima, Giovanna, Huiying,
Kristel, Sam, Simona and Yongmin. I wish also to thank all my friends back home, in
particular Paolo, Dario, Giovanna, Valentina and Irene, for their support.
Last but not least, very special thanks to my parents and my sister for their tireless and
wholehearted support during my studies, which allowed me to arrive until here.
Chapter 1
1
1 β-NAD+: AN ESSENTIAL BIOLOGICAL MOLECULE
1.1 β-NAD+ structure
Nicotinamide adenine dinucleotide, β-NAD+, is a ubiquitous biological molecule found
in all living cells that is involved in many major biological processes. Its structure
consists of two nucleotides, adenosine 5′-monophosphate (AMP) and β-nicotinamide
mononucleotide (β-NMN), joined by a pyrophosphate bond through their phosphate
groups (Figure 1.1).1
Figure 1.1 β-NAD+ structure. AMP (yellow box) and β-NMN (blue box) joined by a pyrophosphate bond.
Its discovery and structure assignment in the 1930s attributed to β-NAD+ the main
function of reversible hydride donor/acceptor because of the ability of the nicotinamide
(NaM) ring to shuttle between the oxidized and reduced form by the transfer of a proton
and two electrons (Figure 1.2). Because of this property, β-NAD+ was classified as a
redox co-substrate for enzymes catalyzing redox reactions. The reduction potential for
the β-NAD+/β-NADH couple, E
0NAD+/NADH, is equal to -0.320 V; therefore, the
oxidation of β-NADH to β-NAD+ is strongly favoured compared to its reduction and
this makes β-NADH a strong reducing agent (Figure 1.2).2
Chapter 1
2
Figure 1.2 β-NAD+/β-NADH redox reaction. The redox reaction can go through a two-electron
mechanism (1, in black) or a one-electron mechanism (1a, 1b, in red), depending on the redox potential of
the coupled substrate, and, in case of enzymatic reactions, on the pH, the polarity of the solvent and steric
requirement.2, 3
Generally, enzymatic reactions proceed via 1, where the enantiotopic protons, pro-R and
pro-S, can be preferentially transferred depending on the enzymes. The one-electron reduction potential
for the couple β-NAD·/β-NADH is E0NAD·/NADH = 0.282 V (1a); for the couple β-NAD
+/ β-NAD·, it is
E0NAD+/NAD· = -0.922 V (1b). The overall reaction (1) generates β-NAD
+ from β-NADH by the loss of a
proton and two electrons (E0
NAD+/NADH = (E0NAD·/NADH + E
0NAD+/NAD·)/2 = -0.320 V).
Besides its redox property, the structure of β-NAD+ is characterized by two high-energy
bonds, the nicotinamide N-glycosidic bond and the pyrophosphate bond, and cleavage
of either releases a substantial amount of energy (Figure 1.3). The cleavage of the
nicotinamide N-glycosidic bond (a) results in the formation of NaM and adenosine
diphosphate ribose (ADPR); on the other hand, cleavage of the pyrophosphate bond (b)
generates β-NMN and AMP. These bonds, extremely labile and unstable, make β-NAD+
easily susceptible to enzymatic and chemical hydrolytic degradation.
The enzymatic cleavage of either high-energy bond in the structure of β-NAD+ is
catalyzed by NAD+-consuming enzymes.
4-7 These enzymes use β-NAD
+ as a substrate
to generate important signalling molecules (e.g. cyclic ADPR, or cADPR, and ADPR),
or covalently modify the acceptors involved in their enzymatic reactions.
Chapter 1
3
Figure 1.3 β-NAD+ and its high-energy bonds. The cleavage of nicotinamide N-glycosidic bond (a)
generates NaM and ADPR; the cleavage of pyrophosphate bond (b) generates β-NMN and AMP.
Regarding its chemical hydrolytic degradation, β-NAD+
is very unstable under basic
conditions, while relatively stable under acidic conditions.8 It has been shown that, at
concentrations below 1 M KOH, the main hydrolytic reaction is the cleavage of the
nicotinamide N-glycosidic bond (Figure 1.3 a, Figure 1.4 a). Indeed, under these
conditions the ionization of the ribose diol (pKa = 11.9) would accelerate the release of
NaM because of the electron-donating ability of the anion that stabilizes the
oxocarbenium ion intermediate; evidence of this effect is given by the stability of O-
isopropylidene protected nicotinamide ribonucleoside (NR) to the cleavage of the
nicotinamide N-glycosidic bond. At concentrations above 1 M KOH, the primary
reaction would become the nucleophilic attack of the hydroxide on the nicotinamide
ring to generate 2-hydroxy-3-pyridinecarboxaldehyde (2HPC) by ring opening,
elimination and reclosing (Figure 1.4 b).9-11
Chapter 1
4
Figure 1.4 β-NAD+ decomposition pathways in basic conditions. a) The cleavage of nicotinamide N-
glycosidic bond with formation of NaM and ADPR occurs at concentration ≤ 1 M KOH. b) The
hydroxylation of NaM, ring opening, release of 1′′-amino adenosine diphosphate ribose (ADPRNH2), ring
closing with formation of 2HPC occurs at concentration > 1 M KOH.
1.1.1 Redox properties of β-NAD+
Owing to its reversible redox activity, β-NAD+(H) is involved in fundamental metabolic
processes occurring in prokaryotes and eukaryotes, such as glycolysis, the Krebs cycle
and cellular respiration (Figure 1.5).1 During the oxidative catabolism of different
substances (e.g. carbohydrates, lipids, proteins) to a common intermediate (acetyl
coenzyme A, or acetyl-CoA), the energy released can be used to reduce β-NAD+ to β-
NADH (endergonic reduction) in the cytosol. β-NADH molecules are then re-oxidized
into β-NAD+ at the surface of the mitochondrial membrane in eukaryotes, generating
electrons that are transferred to the electron transport chain, comprising an enzymatic
series of redox reactions across the mitochondrial inner membrane, to finally reduce O2
to H2O. During the reactions occurring in the electron transport chain, a proton gradient
across the mitochondrial membrane is generated, providing the energy to catalyze the
endergonic synthesis of adenosine triphosphate, ATP, from adenosine diphosphate,
ADP, and phosphate, Pi (oxidative phosphorylation).1
Chapter 1
5
Figure 1.5 Glucose metabolism and enzymatic reactions involving β-NAD+(H) as a redox substrate. a)
refers to processes occurring in the cytosol, b) to the oxidative phosphorylation occurring in eukaryotes at
the interface with the mitochondrial membrane.
glucose
2 pyruvate
glycolysis 2b-NAD+
2b-NADH
2 Acetyl-CoA
2b-NAD+
2b-NADH
Krebscycle
4 CO2
6b-NAD+
6b-NADH
Glyceraldehyde 3-phosphate
dehydrogenase
Pyruvatedehydrogenase
Isocitrate/a-ketoglutarate/malate
dehydrogenase
Oxidativephosphorylation
lipids
proteins 2 CO2
a)
b)
Chapter 1
6
Enzymes using β-NAD+(H) as a redox substrate keep constant the cytosolic
concentration of the dinucleotide that shuttles between its oxidized and reduced form.
The β-NAD+:β-NADH ratio is connected to the intracellular redox state and influences
several metabolic processes; changes in this ratio reflect metabolic changes in the
cells,12
hence the importance to cellular functions to monitor the β-NAD+:β-NADH
ratio.
Intracellular β-NAD+(H) can be present in the free and enzyme-bound form; the total
(free + bound) intracellular β-NAD+:β-NADH ratio fluctuates between 0.1 and 10 in
different organisms, while the cytosolic free β-NAD+:β-NADH ratio is around 500,
since most β-NADH is in its enzyme-bound form.12, 13
The β-NAD+:β-NADH ratio has
been estimated in different cellular compartments, for example in the cytosol and the
mitochondria, showing quite similar values, with the oxidized form favoured to the
reduced form.12
The preference for a substantial free pool of β-NAD+ compared to β-
NADH reflects the requirement of β-NAD+ in the catabolic processes occurring in the
cytosol.
1.1.2 β-NAD+ catabolism and NAD+-consuming enzymes
As mentioned in 1.1.1, the use of β-NAD+(H) as a substrate by redox enzymes does not
alter its concentration in living cells. By contrast, β-NAD+ is used up by NAD
+-
consuming enzymes,7 strictly involved in the catabolism of β-NAD
+.
ADP-ribosyl cyclases (ADPRCs), NAD+-glycohydrolases (NADases), ADP-ribosyl
transferases (ARTs), polyADP-ribosyl polymerases (PARPs) and sirtuins all share the
same substrate/co-substrate, β-NAD+, and catalyze the cleavage of the nicotinamide N-
glycosidic bond with consequent release of NaM as a reaction product. In ADPRCs and
NADases, β-NAD+ is the only enzyme substrate and it is used to generate cADPR and
ADPR (Figure 1.6). In addition, ADPRCs can also catalyze the exchange of the
nicotinamide ring with the nicotinic acid (NA) from β-NAD(P)+, or (2′-phosphate)
nicotinamide adenosine dinucleotide, to generate β-NAAD(P)+, or (2′-phosphate)
nicotinic acid adenosine dinucleotide, not consuming the β-NAD+ scaffold. All these
molecules have been shown to be involved in Ca2+
signalling, although through
Chapter 1
7
completely different pathways (Figure 1.6).4, 6, 7, 13
ARTs and PARPs use β-NAD+ as a
co-substrate in their reactions, where they catalyze respectively the transfer of one unit
of ADPR, or many branched polymers of ADPR, to amino acid protein residues (Figure
1.6). Protein modification by ARTs would have a regulatory and signalling role, acting
by inhibition of the normal activities of the protein acceptors. Regarding PARPs, a
variety of these enzymes have been found in mammals; the best studied is PARP-1, for
which the modification of the protein is triggered by breaks in the DNA and, therefore,
it is involved in the DNA repair response and transcriptional activation. Another
important PARP is tankyrase-1, which regulates the maintenance of telomere length
(Figure 1.6).4, 6, 7
Finally, sirtuins use β-NAD+ as a co-substrate to mainly catalyze the
deacetylation of proteins and generate NaM and 2′′(3′′)-acetate-ADPR, or 2′′(3′′)-OAc-
ADPR (Figure 1.6). The deacetylation of proteins is a fundamental regulatory biological
process; hence, the interest toward sirtuins has increased since they appear to be
involved in epigenetic, ageing and several human diseases.5, 6
In addition, the reaction
product, 2′′(3′′)-OAc-ADPR, seems to be a signalling molecule involved again in Ca2+
release.6, 7, 13
Sirtuins could also have a secondary activity as ARTs, but it is still not
clear in which conditions this activity would be stimulated (Figure 1.6).14
Nucleotide pyrophosphatases (NPPs) and NAD+-dependent DNA ligases use β-NAD
+
respectively as a substrate and a co-substrate.15, 16
Indeed, both enzymes catalyze the
cleavage of the pyrophosphate bond to generate AMP and β-NMN (Figure 1.6); in the
case of the DNA ligases, this reaction is in conjunction with the ligation of a single
strand DNA break. With respect to AMP generated from NPPs, which are generally
ecto-enzymes with the catalytic site in the extracellular space, this can be recycled by
ecto-5′-nucleotidases to obtain adenosine, involved in purinergic signalling (Figure
1.6).17
The other reaction product, β-NMN, is not a direct precursor of β-NAD+ in its
biosynthesis, as will be shown in 1.1.3; nevertheless, there is evidence for the existence
of enzymes that would generate precursors of β-NAD+ from β-NMN, such as NR or
NaM.18, 19
These enzymes have not been well characterized yet in humans but highlight
the existence of many pathways for the biosynthesis of β-NAD+.
Chapter 1
8
Figure 1.6 NAD+ derivatives from NAD
+-consuming enzymes activities and their functions. Ca
2+-
signalling messengers in the blue boxes; purinergic signalling messenger from AMP, and β-NAD+
precursors from β-NMN in the green box; regulatory protein modifications (e.g. DNA repair,
transcription activation, ageing) in the red box. The overlap between the boxes indicates the strong
relations occurring between the activities of all these enzymes. Abbreviations used: NAD+, nicotinamide
adenine dinucleotide; ADPR, adenosine diphosphate ribose; cADPR, cyclic ADPR; cADPRP, 2′-
phosphate cADPR; NADP+
, 2′-phosphate NAD+; NAADP
+, 2′-phosphate nicotinic acid adenine
dinucleotide; 2′′(3′′)-OAc-ADPR, 2′′(3′′)-acetate-ADPR; ADPRC, ADPR cyclase; NADase, NAD+-
glycohydrolase; NADK, NAD+ kinase; NPP, nucleotide pyrophosphate; PARP, polyADPR polymerase;
ART, ADPR transferase; SIRT, sirtuins; DNA Lig, DNA ligase.
Extremely important among the enzymes that transform but do not consume the β-
NAD+ scaffold is the NAD
+ kinase (NADK); this enzyme catalyzes the phosphorylation
of β-NAD+ on 2′-OH to give β-NADP
+ (Figure 1.6).
13 In turn, two other fundamental
Ca2+
-signalling messengers, β-NAADP+ and cADPRP (2′-phosphate cyclic ADPR), can
be generated from β-NADP+ by the action of ADPRCs (Figure 1.6).
β-NADP+, having the same reduction potential as β-NAD
+, can undergo reduction on
the nicotinamide ring to obtain β-NADPH. The concentration of the two dinucleotide
couples is regulated by set of different enzymes; in particular, the concentration of β-
NADP+ is generally much lower than β-NAD
+ for similar concentrations of β-NADPH
and β-NADH (Table 1.1). Therefore, the β-NADP+:β-NADPH ratio is in favour of the
b-NAD+
cADPR
ADPR
2’’(3’’)-OAc-ADPR
b-NAADP+ cADPRP
b-NADP+
Acceptor-ADPR
Acceptor-polyADPR
b-NMN + AMP
b-NAAD+
ADPRC
Chapter 1
9
reduced form, in contrast to what is observed for the β-NAD+:β-NADH ratio.
13, 20 This
fact reflects the importance of β-NADP+(H) on the balance of the intracellular redox
state in response to cellular oxidative stress; indeed, the substantial free pool of β-
NADPH compared to β-NADH is needed to react and inactivate the reactive oxygen
species (ROS), that are responsible for oxidative stress and, therefore, a significant
cause of cellular damages.
Metabolite Concentration [M]
β-NAD+ 2.6 x 10
-3
β-NADH 8.3 x 10-5
β-NADP+ 2.1 x 10
-6
β-NADPH 1.2 x 10-4
Table 1.1 Intracellular metabolite concentrations in E. coli.20
When extensive DNA damage occurs in cells, it triggers a cascade of enzyme activities
responsible for DNA repair; PARP-1, one of the most important DNA-repair enzymes
in mammals, begins to produce very long ADPR branches as part of its function, almost
completely depleting the intracellular free pool of β-NAD+, and substantially altering
the intracellular β-NAD+:β-NADH ratio. This change would dramatically reflect on the
intracellular lack of β-NADH, hence in ROS protection; therefore, this underscores the
importance of the free pool of β-NADPH to respond to cellular oxidative stress.13
Moreover, the different ratios of β-NAD+(H) and β-NADP
+(H) reflect another important
difference between the two dinucleotide couples. While β-NAD+(H) is mostly in its
oxidized form, hence it is used in oxidative catabolic reactions, β-NADP+(H), mostly
present in its reduced form, is almost exclusively used in reductive anabolic reactions,
where complex substances (e.g. carbohydrates, lipids, proteins) are formed by reductive
steps from acetyl-CoA (Figure 1.7). This is a fascinating example of how nature has
developed independent catabolic and anabolic pathways depending respectively on
similar dinucleotides exploiting differently their reversible redox activity.1
Chapter 1
10
Figure 1.7 Catabolic processes (blue arrow) mainly involving the reduction of β-NAD+ to β-NADH;
anabolic processes (red arrow) involving the oxidation of β-NADPH to β-NADP+.
1.1.3 β-NAD+ biosynthesis
Since redox enzymes keep the β-NAD+(H) concentration constant while NAD
+-
consuming enzymes use it in their reactions and decrease its concentration, β-NAD+(H)
homeostasis in living cells is maintained by two different biosynthetic pathways that
regenerate β-NAD+. These pathways are classified as the de novo pathway, in which β-
NAD+ is generated from tryptophan (Trp) by different enzyme-catalyzed steps, and the
salvage pathway, where β-NAD+ is formed from recycling products, such as NaM, NA,
also known as niacin or vitamin B3, and NR, which come from the depletion of β-
NAD+ or are introduced by diet into the cells (Figure 1.8).
18, 21 The two different
pathways converge at a certain point and share the same enzymes to give the final β-
NAD+.
Different organisms show different preferences towards the de novo and salvage
pathway; generally, the salvage pathway is preferred to the de novo one. In addition,
different organisms exhibit clear preferences toward salvage pathway precursors: for
instance, yeast prefers NA to NaM, since it expresses the enzyme nicotinamide
deamidase, which deamidates NaM to NA, which is not present in mammals. In yeast
and bacteria, another salvage precursor can be NR, which can be converted into β-NMN
by NRK (NR kinase).
anabolism
AcetylCoACarbohydrates, lipids, proteins
catabolism
b-
b- b-
b-
Chapter 1
11
In the salvage pathway of mammals, the precursor NaM is directly transformed to β-
NMN by the action of NaMPRT (NaM phospho-ribosyl transferase), which undergoes
the coupling with AMP by NMNAT (NMN adenylyl transferase) to give β-NAD+. In
yeast, NA, obtained from NaM by nicotinamide deamidase, is transformed in nicotinic
acid mononucleotide (NAMN) by NAPRT (NA phospho-ribosyl transferase); then,
NAMN is coupled to AMP by NAMNAT (NAMN adenylyl transferase) to give β-
NAAD+, which is amidated by NAD
+-synthetase to finally give β-NAD
+ (Figure 1.8).
Figure 1.8 β-NAD+ biosynthesis and metabolism. De novo pathway (red box); salvage pathway (blue
boxes); products from NAD+ metabolism (green boxes). Abbreviations used: Trp, tryptophan; QA,
quinolic acid; QAPRT, QA phospho-ribosyl transferase; NA, nicotinic acid; NaM, nicotinamide; NAMN,
nicotinic acid mononucleotide; NMN, NaM mononucleotide; NR, NaM ribonucleoside; NRK, NR kinase;
N(A)MNAT, NaM (NA) mononucleotide adenylyltransferase; NAAD+, NA adenine dinucleotide; NAD
+,
NaM adenine dinucleotide; NAPRT, NA phospho-ribosyl transferase; NaMPRT, NaM phospho-ribosyl
transferase; NADK, NaM adenine dinucleotide kinase; NADP+, 2′-phosphate NaM adenine dinucleotide;
ADPR, adenine diphosphate ribose; ADPRC, ADPR cyclase; cADPR, cyclic ADPR; NPP, nucleotide
pyrophosphatase; AMP, adenosine monophosphate; SIRT, sirtuins; 2′′(3′′)-OAc-ADPR, 2′′(3′′)-acetate-
ADPR; NADase, NAD+-glycohydrolase; ART, ADPR transferase; PARP, polyADPR polymerase.
Chapter 1
12
Therefore, each organism has a similar set of enzymes (e.g. phospho-ribosyl
transferases and adenylyl transferases) to perform the salvage pathway but with
specificity depending on the favoured precursor. Structural studies of the active site of
mammalian, yeast and bacterial enzymes involved in salvage pathways have shown
substantial differences responsible for the different substrate specificity in spite of their
same function; hence, although the sequence of reactions is conserved among different
organisms, an evolutionary divergence has allowed them to differentiate the pathway
specificity.22
In the human salvage pathway the rate-limiting enzyme is NaMPRT, which catalyzes
the phosphoribosylation of NaM from 5-phosphoribosyl-1-pyrophosphate with release
of β-NMN and pyrophosphate, PPi. This enzyme is crucially involved in the β-NAD+
biosynthesis and, therefore, in the activity of NAD+-consuming enzymes, responsible
for many biological processes involved in human diseases; hence, there is an increasing
interest toward NaMPRT as a therapeutic target.21, 23
Less crucial, but still important, is
NMNAT, which catalyzes the adenylation of β-NMN from ATP with release of β-
NAD+ and PPi; also the control of its activity could potentially influence the metabolism
of β-NAD+ and, therefore, have an impact on the cure of some human diseases.
21
In mammals, NA is essential due to its role as a precursor of β-NAD+. It can be obtained
by diet or synthesized from Trp; consequently, lack of NA or Trp in the diet can cause
pellagra in humans, a deficiency disease. Hence, the de novo pathway is fundamental to
provide NA from Trp and β-NAD+ in mammals in the case of restricted NA availability.
The de novo pathway synthesizes β-NAD+ starting from Trp through the kynurenine
pathway (Scheme 1.1).18
The opening of the pyrrole ring of Trp in conjuction with its
oxidation, catalyzed by indoleamine-2,3-dioxygenase (IDO) in most tissues and
tryptophan-2,3-dioxygenase (TDO) in the liver, generates formylkynurenine; the latter
is hydrolysed to L-kynurenine by kynurenine formamidase (KFase). L-Kynurenine
undergoes hydroxylation in position 3 by kynurenine-3-hydroxylase (K3H) to give 3-
hydroxykynurenine, which is converted to 3-hydroxyanthranilic acid and L-alanine by
kynureninase (kyase). Then, 3-hydroxyanthranilate-3,4-dioxygenase (3HAO) catalyzes
the opening of the phenyl ring and its oxidation by insertion of molecular oxigen to give
2-amino-3-carboxy muconate semialdehyde, which spontaneously cyclizes into
Chapter 1
13
quinolinic acid (QA). The latter is transformed to NAMN by the action of quinolinate
phospho-ribosyl transferase (QAPRT) (Figure 1.8); hence, the de novo pathway
converges on the salvage pathway to finally give β-NAD+.
The rate-limiting steps in the de novo pathway are the reactions catalyzed by IDO/TDO
and QAPRT; the rates of these reactions influence the rate of the de novo pathway and
strictly forces it toward the conjunction with the salvage pathway.18
IDO and K3H have
shown to be potential therapeutic targets toward a wide class of human diseases.21
Scheme 1.1 De novo pathway. Abbreviations used: L-Trp, L-tryptophan; IDO, indoleamine-2,3-
dioxygenase; TDO, tryptophan-2,3-dioxygenase; KFase, kynurenine formamidase; K3H, kynurenine-3-
hydroxylase; kyase, kynureninase; 3HAO, 3-hydroxyanthranilate-3,4-dioxygenase; N.E., not enzymatic.
1.2 NAD+-consuming enzymes
The main NAD+-consuming enzymes, their activities and reaction products have been
highlighted in 1.1.2. Since the main product released from the activity of NAD+-
consuming enzymes is NaM, the principal salvage precursor for the biosynthesis of β-
NAD+ in mammals (Figure 1.8), it is clear that a strong relationship exists between the
activity of these enzymes and the biosynthesis of β-NAD+; hence, every enzymatic
Chapter 1
14
reaction and its substrates and products are connected in a delicate equilibrium which
controls the homeostasis of such an essential molecule as β-NAD+.4-7, 18
Herein, the attention will be focused on the relationship existing between the activities
of NAD+-consuming enzymes, discussing how they influence each other and the
extracellular and intracellular homeostasis of β-NAD+. Later, this chapter will present a
short review of the existing synthetic NAD+ derivatives and their use as inhibitors for
drug therapies, or biochemical tools for the development of assays to monitor the
activity of NAD+-consuming enzymes. In this context, the project and its objectives will
be introduced as an alternative and innovative solution to solve biochemical questions
involving the NAD+-consuming enzymes.
1.2.1 NAD+-consuming enzymes: similarities and relationships
The majority of NAD+-consuming enzymes catalyze the cleavage of the nicotinamide
N-glycosidic bond with consequent release of NaM. The cleavage of this bond is an
exergonic reaction;24
in particular, the energy released is equal to -8.2 kcal·mol-1
(for
comparison, the cleavage of ATP to generate ADP and Pi,25
a very well known
exergonic reaction occurring in biochemical processes, is equal to -7.3 kcal·mol-1
).
Within cells, exergonic reactions are generally coupled with endergonic processes but,
in the case of NAD+-consuming enzyme activities, it is not always clear what the energy
released is used for. Indeed, the reaction products from their activities do not keep the
energy released under the form of high-energy bonds, as often occurs in reactions using
ATP; moreover, the catalytic mechanisms of NAD+-consuming enzymes do not justify
the release of such a large amount of free energy.
NAD+-consuming enzymes catalyzing the cleavage of the nicotinamide N-glycosidic
bond have in common an NAD+-glycohydrolase activity; then, other enzymatic
reactions can be coupled to this common one, for instance the cyclization of ADPR to
cADPR (ADPRCs), the transfer of ADPR to protein acceptors (ARTs, PARPs), and
protein deacetylation (sirtuins). The cleavage of the nicotinamide N-glycosidic bond by
these enzymes occurs via a two-step mechanism resembling its chemical hydrolysis,
where the ionization of the ribose diol stabilizes the formation of an oxocarbenium
Chapter 1
15
intermediate accelerating the release of NaM (Figure 1.4). In the enzymatic catalysis,
the interaction of 2′′-OH and 3′′-OH of the nicotinamide ribose with a carboxylic
residue in the active site, would form a partial negative charge on the diol (Figure 1.9).
The latter would exert an electron-donating effect on the ribose ring destabilizing the
nicotinamide N-glycosidic bond and, therefore, favouring the release of NaM and the
formation of an oxocarbenium intermediate.8 This intermediate, stabilized on the
nicotinamide ribose by H-bonds, would then undergo different fates.
Figure 1.9 Formation of an oxocarbenium intermediate during the NAD+-glycohydrolase activity of
NAD+-consuming enzymes. A carboxylic acid residue stabilizes the oxocarbenium intermediate by H-
bonds, which can evolve afterwards into different compounds by nucleophilic attack on C-1′′.
ADPRCs, NADases, PARPs, and ARTs mechanisms all proceed via the hypothetical
oxocarbenium intermediate, as demonstrated by the presence in each of these enzymes
of highly conserved glutamate residues.8, 26-30
Still controversial is the mechanism
adopted by sirtuins; indeed, it could involve an SN1 reaction via the oxocarbenium
intermediate,31
or an SN2 reaction with concurrent nucleophilic attack and elimination of
NaM.32
The energy released by the cleavage of the nicotinamide N-glycosidic bond is
temporarily conserved in the highly reactive oxocarbenium intermediate; however, the
following transformations in which it is involved are not very endergonic reactions. For
instance, in ADPRCs the cyclization of the intermediate into cADPR requires the
adenine ring to turn from anti into syn conformation (Chapter 4, see 4.2.3);26
in
NADases, the intermediate can undergo nucleophilic attack by water (hydrolysis) to
form ADPR, or by other nucleophiles to undergo an exchange reaction;24
in ARTs and
Chapter 1
16
PARPs the intermediate reacts with the nucleophilic residues of a protein acceptor to
give an ADP-ribosylated acceptor;28
and, finally, in sirtuins the eventual oxocarbenium
intermediate undergoes acetylation on 2′′(3′′)-OH to give 2′′(3′′)-OAc-ADPR and the
corresponding deacetylated protein.31
Therefore, the energy conserved in the
oxocarbenium intermediate is thought in part to be used for the former transformations,
even if not highly endergonic, but also to be coupled to other biological processes, not
immediately connected to the enzyme catalytic mechanisms. Hence, in the human
ADPRC, CD38, it has been shown the necessity for this ecto-enzyme, that binds β-
NAD+ from the extracellular space, to dimerize to actively produce cADPR and
transport it inside the cells (Figure 1.12); this process would require energy, which
could come from the cleavage of the nicotinamide N-glycosidic bond.33
Furthermore, in
PARPs the formation of ADPR branches as post-transcriptional modification is required
as a signal to actively recruit other proteins involved in DNA repair, and energy would
be needed in this process;7 in addition, but not yet fully demonstrated, the ADPR
branches could be used as a source to synthesize ATP.34
Finally, in sirtuins the
deacetylation of histones, proteins required for the compacting of chromatin, would be
coupled to chromatin condensation, an active process for which the energy released
could be used.25
In contrast to ADPRCs, NADases, PARPs, ARTs and sirtuins, NPPs and NAD+-
dependent DNA ligases catalyze the exergonic cleavage of the β-NAD+ pyrophosphate
bond; in both enzymes, the energy released is conserved by the formation of active
enzyme-AMP complexes in the first step of their catalytic mechanisms. Afterwards,
NPPs release AMP with an apparent loss of free energy (see 4.1, Scheme 4.1); by
contrast, DNA ligases transfer the energy from the active enzyme-AMP complex to the
active DNA-AMP complex, to eventually form the phosphodiester bond on the nicked
DNA strand (see 5.1, Scheme 5.1). In the case of NPPs, the apparent loss of energy
could be coupled to more complex biological processes, such as the formation of
adenosine from AMP and, therefore, be linked to its role in purinergic signalling.17
Chapter 1
17
It is interesting to study the complex relationship existing between the different NAD+-
consuming enzymes and how they influence major biological processes connected to
many human diseases; a summary of the main effects exerted by these connections will
be provided. An aspect shared by all those enzymes with NAD+-glycohydrolase activity
(ADPRCs, NADases, PARPs, ARTs and sirtuins) is the inhibitory effect exerted by
NaM, their reaction product, on their activities. High concentrations of NaM from the
activity of any NAD+-consuming enzyme inhibit the activities of all the others; hence,
the depletion of NaM via the salvage β-NAD+ biosynthetic pathway
5, 18 is fundamental
(Figure 1.10). In mammals, the activity of PARPs and CD38,35
having both ADPRC
and NADase activities, is responsible for the depletion of large amounts of β-NAD+ at a
very fast rate.7 Therefore, NaM produced by these enzymes can strongly inhibit sirtuins,
which activity can be re-established by removal of NaM through the action of NaMPRT
in mammals,23
or NaM deamidase (PNC1) in yeast7 to form precursors of β-NAD
+.
Since sirtuins are proposed to be involved in many critical biological processes
(glucose, fat and insulin metabolism, ageing and life span increase, DNA transcription
and repair, apoptosis), the regulation of their activity seems crucial; hence, it is
important to develop sirtuins inhibitors/activators, or CD38/PARPs inhibitors to block
the depletion of β-NAD+ that inhibits sirtuins.
35 Since the activities of all these enzymes
are important in cells, the choice of the process to block with a drug is crucial to avoid
possible side effects.
Figure 1.10 β-NAD+ metabolism and its influence on the activity of sirtuins. Abbreviations used: NAD
+,
nicotinamide adenine dinucleotide; NaM, nicotinamide; PARP, polyADPR polymerase; NaMPRT, NaM
phospho-ribosyl transferase.
b-NAD+ NaM sirtuins inhibitionsirtuins activation
CD38,PARPs
NaMPRT (mammals)NaM deamidase (yeast)
Chapter 1
18
Sirtuins, PARPs and DNA ligases are intracellular enzymes involved in DNA repair
processes (Figure 1.11).36
When DNA damages occur in the cells, PARP-1 starts
immediately its ADP-ribosylation or that of other proteins as a signal to trigger the
recruitment to the DNA break of all proteins involved in DNA repair process (1). The
formation of branches of ADPR, up to 200 units in length, occurs in a very short time
and almost completely depletes the free intracellular β-NAD+ pool with a huge increase
of NaM concentration in the proximity of the DNA break.7 The latter affects the activity
of sirtuins, strongly inhibited by NaM (2), which therefore are not able to deacetylate
histones (3).5 Therefore, the DNA is unwound because of the fully acetylated histones
in chromatin and the proteins for the DNA repair, including DNA ligases, can move
close to the DNA and interact with it to fix its damage. When ADPR branches have
reached the critical length, PARP-1 decreases its activity and, consequently, the NaM
concentration near to the DNA break decreases as well; then, sirtuins can deacetylate
histones re-compacting again the chromatin (4).36
However, the inhibition of sirtuins
can affect the activity of other biological processes. For instance, sirtuins can
deacetylate p53, a protein responsible for DNA repair or apoptosis when the DNA
damage is irreparable; p53 acetylation (7), in the case of the inhibition of sirtuins,
corresponds to its activation and, therefore, to cell apoptosis (8). On the other hand,
sirtuins can deacetylate FOXO transcription factors, involved as well as p53 in
apoptotic processes; oppositely to the effect exerted on p53, FOXO acetylation (5), in
the case of the inhibition of sirtuins, corresponds to their deactivation and, therefore, to
cell survival (6).
Moreover, a PARP-overstimulation leads directly to cell apoptosis (9) because of the
complete depletion of the intracellular β-NAD+ pool, which cannot be formed in a short
time by the β-NAD+ biosynthetic pathways. Conversely, the inhibition of PARPs can be
connected to cell survival (18), despite the loss in DNA repair response, because it leads
to the activation of sirtuins (11). The latter effect can in turn lead to apoptosis or cell
survival depending on the activity exerted respectively on FOXO transcription factors
(14) or p53 (16)/histones (12). Hence, the inhibition/activation of sirtuins by PARP
activity can result in completely different cellular fates depending on the main cellular
Chapter 1
19
responses and their activation times compared to the DNA repair processes (Figure
1.11).36
Figure 1.11 Inhibition/activation of sirtuins by the activity of PARPs in response to DNA damage. Cells
can undergo different fates (1-18) depending on the related activities of sirtuins and PARPs and their
modes of action. FOXO (Forkhead box proteins, class O) are involved in cellular proliferation, lifespan
and stress tolerance. p53 (protein 53) is a tumor suppressor.
DNA damage
PARP inhibitionPARP activation
PARP inhibitors
SIRT activationSIRT inhibition
Nicotinamide
p53 Acetylation/
activation
FOXOAcetylation/
inhibition
Histoneacetylation
p53 Deacetylation/
inhibition
FOXODeacetylation/
activation
Histonedeacetylation
DNA REPAIR/SURVIVALAPOPTOSIS
1 10
9 18
2
3
4
75
6
8
11
12
13
14
15
16
17
Chapter 1
20
1.2.2 Intracellular and extracellular β-NAD+ balance
A delicate balance exists between extracellular and intracellular β-NAD+ concentration
in multicellular organisms to regulate the functions of such a fundamental biological
molecule.37
It has been estimated that in multicellular organisms extracellular β-NAD+
concentration is normally in the order of submicromolar range (10 – 50 nM), much
lower than the intracellular one (0.1 – 3 mM).7 These concentrations are the result of the
activities of an intricate network of nucleoside/nucleotide receptors and NAD+-
consuming enzymes that keep a constant extracellular balance between β-NAD+ and its
degradation products, regulating the release of intracellular β-NAD+ and its degradation
outside the cells (Figure 1.12).38, 39
The main cause of intracellular β-NAD+ release and, consequently, extracellular β-
NAD+ increase is cellular apoptosis and lysis. In addition, β-NAD
+ transporters
(Connexin 43 transmembrane hemi channels) have recently been identified as
responsible for the influx/efflux of β-NAD+ into/outside the cells, allowing the
regulation of its concentration in response to different biological processes, and to
modulate intercellular communications.40
Indeed, it has been shown that extracellular β-
NAD+ concentration substantially increases during inflammatory responses, since β-
NAD+ can act as a cytokine that triggers signalling cascade responses through its
binding to specific purinergic receptors. For instance, its binding to P2Y11 causes an
elevated intracellular Ca2+
release and cAMP (cyclic AMP) changes leading to anti-
inflammatory responses (Figure 1.12).37, 39, 41
Extracellular β-NAD+ homeostasis is further controlled by its degradation through the
activities of some NAD+-consuming enzymes, for instance mammalian ADPRC
(CD38), NADases, ARTs and NPPs. All these enzymes belong to the ecto-enzyme
class, characterized by the presence of their catalytic active site in the extracellular
space. As discussed in 1.1.2, they catalyze the cleavage of extracellular β-NAD+ to
generate signalling molecules (e.g. cADPR, ADPR, β-NAADP+) or protein ADP-
ribosyl modifications that potentially act in the intracellular space. Hence, it emerges the
paradox of these enzymes, that are active in the extracellular space but involved in
intracellular processes (e.g. Ca2+
release, post-transcriptional modifications, regulation
Chapter 1
21
of intracellular sirtuins and PARPs activities).7 The first issue in solving this paradox is
due to the low extracellular β-NAD+ concentration, below the Km of NAD
+-consuming
enzymes, which makes their activity rate very low. However, this allows to modulate
finely their activity, since even small increases in extracellular β-NAD+ concentration
by the action of β-NAD+ transporters,
40 favouring the release of intracellular β-NAD
+ in
the extracellular space, could dramatically increase the activity of ecto NAD+-
consuming enzymes.
Another issue is the connection between the extracellular activity of these enzymes and
the triggered intracellular processes. In the case of the cyclase activity of CD38, mostly
present as an ecto-enzyme, the extracellular β-NAD+ is bound into the catalytic site to
generate cADPR, which stimulates intracellular Ca2+
release from the endoplasmic
reticulum. The discovery of the dimerization and internalization of CD38 to catalyze the
enzymatic reaction and actively transport the product inside the cell explained the link
between the extracellular β-NAD+ binding site and the intracellular product release.
33
Moreover, specific cADPR-transporters have been found in the cellular membrane,
responsible for the intracellular uptake of cADPR.42
Regarding the NADase activity of
CD38, it generates extracellular ADPR, which promotes intracellular Ca2+
release41
via
pathways mediated by IP3 (inositol 1,4,5-triphosphate), possibly binding to the
extracellular P2Y purinergic receptor.43
This process is completely independent from
the Ca2+
-permeable TRPM2 channel transport, which mediates intracellular Ca2+
release
from internal compartments and/or Ca2+
influx from the extracellular space by the
action of second messengers, such as intracellular ADPR (produced by degradation of
ADPR branches), cADPR, 2′′(3′′)-OAc-ADPR (produced by sirtuins) and β-NAADP+.43
The extended activation of this non-selective cation channel causes cell death. β-
NAADP+, generated by the exchange activity of CD38 from extracellular β-NADP
+, can
be, in turn, transported inside the cells via specific β-NAADP+ transporters (Figure
1.12).44
Mammalian ARTs have been discovered quite recently and most of them are classified
as ecto-enzymes, therefore acting in the extracellular space and generating ADP-ribosyl
modifications of extracellular proteins; in particular, they are involved in receptor
modifications and triggering signalling cascades during immune and inflammatory
Chapter 1
22
responses.7 For instance, in mouse T-cells the ADP-ribosylation of the purinergic
receptor P2X7 by ART2 induces its activation and a cascade of signalling processes,
finally leading to cell death.37
A large number of intracellular proteins undergo ADP-
ribosylation, which cannot be ascribed to ecto-ARTs; therefore, the existence of
intracellular ARTs or the secondary ART activity of sirtuins14
would be responsible for
intracellular modifications (Figure 1.12).7
Finally, NPPs, as well as the previous enzymes, belong to the ecto-enzymes class. The
presence of NPPs on the cellular surface is fundamental for the recycling of
extracellular nucleotides and for keeping the balance between nucleotides/nucleosides.
For instance, the CD38-generated ADPR can be transformed by NPPs into AMP, which
in turn can be hydrolysed into adenosine (Ado), which is able to bind to purinergic
receptors and stimulate signalling responses (Figure 1.12).
Figure 1.12 Pathways for the extracellular/intracellular β-NAD+ balance. Abbreviations used: NAD
+,
nicotinamide adenine dinucleotide; ADPR, adenosine diphosphate ribose; cADPR, cyclic ADPR;
NAADP+
, 2′-phosphate nicotinic acid adenine dinucleotide; 2′′(3′′)-OAc-ADPR, 2′′(3′′)-acetate-ADPR;
AMP, adenosine monophosphate; Ado, adenosine; Cx43, Connexin 43; ART2, ADPR transferase 2; NPP,
nucleotide pyrophosphatase.
All these ecto-enzymes are therefore involved, not only in the homeostasis of
extracellular β-NAD+, but in fundamental signalling pathways, where activation triggers
anti-inflammatory and immunosuppressive responses. The increased extracellular β-
Chapter 1
23
NAD+ concentration in connection with increased expression of NAD
+-consuming ecto-
enzymes and purinergic receptors is the effect observed following cellular stress at the
beginning of inflammatory responses; tumor cells seem to modulate these interactions
to escape the immune response.39
Intracellular β-NAD+, as already shown in 1.1.3, is synthesized in the cytosol by the
salvage or de novo pathways and used by intracellular enzymes, such as redox (NAD+-
dependent dehydrogenases) and NAD+-consuming enzymes (PARPs, sirtuins, NAD
+-
dependent DNA ligases). Some of these enzymes perform their crucial activities inside
the mitochondria, where an appropriate β-NAD+:β-NADH ratio is fundamental for a
variety of biological processes; however, for a long time it was thought that β-NAD+(H)
was unable to permeate the mitochondrial membrane. Recently, a mitochondrial β-
NAD+ transporter, Ndt1p, has been identified in Saccharomyces cerevisiae as a specific
β-NAD+ carrier. It imports β-NAD
+ into the mitochondria preferentially by exchange
with mitochondrial deoxy-AMP (dAMP) and deoxy-GMP (dGMP), and to a small
extent by uniport reaction (Figure 1.12).45
1.3 NAD+ derivatives as mechanistic probes for NAD+-dependent enzymes
As shown in the previous sections, NAD+-dependent enzymes play a fundamental role
in many biological processes and alterations and modifications to their normal status are
linked to several human diseases. Therefore, the necessity to investigate and, eventually,
modulate or block NAD+-dependent enzymes has led to the development of NAD
+
derivatives that mimic the natural substrate (NS) and act either as non-natural substrates
(NNSs) for biochemical applications or potential enzyme inhibitors for drug discovery.
In general, designing derivatives of the natural substrates requires minor modifications
of the natural molecule scaffold in order to keep intact the essential groups interacting
with the enzymes and allowing its binding to the catalytic sites. Then, depending on the
modifications, it is possible to generate NNSs, that are metabolized with the same
fashion as the NS, and are consequently its competitive inhibitors; or, simply, enzyme
inhibitors, if the modifications allow their binding but not their metabolism by the
enzymes. In the development of NAD+ derivatives it is therefore necessary to look at
the natural β-NAD+ structure and identify the chemical groups that can be subjected to
Chapter 1
24
modifications; then, depending on the need of inhibitors or NNSs, the right
modifications have to be considered.
In the design of inhibitors, if the enzyme structure and mechanism are well known, the
modifications are aimed to block essential groups for the catalytic reactions. In contrast,
in the design of NNSs, the modifications are needed to investigate and monitor the
enzyme structure and/or the related mechanism, without consequently affecting the
essential interactions responsible for the enzymatic catalytic mechanisms. For instance,
an ideal NNS should have an enzyme affinity similar or higher compared to the NS (Km
NNS ≤ Km NS) to bind properly in the active site. In addition, a kcat (turnover number)
lower than the NS (kcat NNS < kcat NS) would allow a slower metabolism of NNS by the
enzyme, permitting in some cases easier kinetic studies and mechanism visualizations
without altering the reaction mechanism itself. By definition, NNSs are competitive
inhibitors of the NS; however, if their Km and kcat are similar to the NS, the enzyme
activity will not be inhibited overall, since both substrates can be used with the same
efficacy. If their Km and kcat are lower (higher affinity and slow turnover) compared to
the NS, then the final effect will be a slowdown of the enzymatic reaction and,
therefore, inhibition; hence, under such conditions, NNSs can be potentially used as
enzyme inhibitors or modulators, or at least give the basis for the design of new
inhibitors. In addition, in the design of NNS, the labelling of the natural molecule
scaffold by groups monitored by sensitive techniques, such as fluorescence, UV
absorbance and immunoanalytical methods, can be extremely useful for the
development of assays that are more sensitive compared to those using the NS. The
labelling of nucleosides/nucleotides with the purpose to generate fluorescent probes will
be discussed in detail in Chapter 3, where the attention will be focused on the
nucleobase substitutions; however, the ribose or phosphate conjugation with
fluorophores by linkers,46-48
where the length is crucial to avoid steric hindrances with
the substrate, is a common labelling method. Also, the introduction of biotin on the
natural molecule scaffold represents a useful labelling method, allowing very sensitive
detections and selective purification due to the strong affinity of biotin and
streptavidin.49
Chapter 1
25
The structure of β-NAD+ is characterized by a variety of functional groups, that are
involved differently in enzyme recognitions and activities; therefore, their modifications
lead to different responses to different enzymes (Figure 1.13).
Figure 1.13 Relationship between group modifications on β-NAD+ structure and specific enzyme
responses. NaM (green box) and nicotinamide ribose (purple box) modifications lead to the inhibition of
dehydrogenases (DHs) and enzymes with NAD+-glycohydrolase activity (ADPRCs, NADases, ARTs,
PARPs, sirtuins). Pyrophosphate bond (yellow box) modifications lead to the inhibition of nucleotide
pyrophosphatases (NPPs). Adenine (blue box) modifications lead to the inhibition of sirtuins (SIRT) and
NAD+-dependent DNA ligases.
Dehydrogenases (DHs), which use the nicotinamide ring for its redox activity, are
mainly inhibited by its modification. NAD+-consuming enzymes that share the NADase
activity (ADPRCs, NADases, ARTs, PARPs, sirtuins, see 1.2.1) are especially inhibited
by NaM/NR modifications, since the formation of an oxocarbenium intermediate is
crucial for their catalytic mechanisms (Figure 1.9). NPPs are mainly inhibited by
pyrophosphate bond modifications, since they catalyze its cleavage. Finally, NAD+-
dependent DNA ligases and sirtuins, which closely interact with the adenine moiety of
β-NAD+ in their catalytic sites (see Chapter 5, respectively 5.1 and 5.2), can be inhibited
by adenine modification in particular positions. By aiming for selective inhibitions
following specific modifications, hypothetically it should be possible to design NNSs of
NAD+-dependent enzymes by modifying the chemical groups not involved in their
inhibition.
NPPs inhibitors
DHs/NADasesinhibitors
SIRT/DNA Liginhibitors
Chapter 1
26
1.3.1 NAD+ derivatives as inhibitors of NAD+-dependent enzymes
1.3.1.1 Nicotinamide modification
NaM is an essential component of β-NAD+, required for its recognition/binding to
several enzymes as well as for their activities. Indeed, the nicotinamide portion is
responsible for the β-NAD+ redox activity, hence for its use as a co-substrate by DHs; in
addition, its positive charge makes the nicotinamide N-glycosidic bond extremely labile,
allowing the use of β-NAD+ as a substrate/co-substrate of NAD
+-consuming enzymes.
The modification of NaM is effective in generating inhibitors, NAD+-like or not, for
NAD+-dependent DHs and NAD
+-consuming enzymes. A variety of studies aimed at
the discovery of inhibitors for important DHs have been focused on the nicotinamide
substitution within the β-NAD+ structure with a variety of aryl/heteroaryl rings (Figure
1.14, benzamide 1, C-nicotinamide 2, picolinamide 3, isonicotinamide 4, thiazole-4-
carboxamide 5, and many others). These main substitutions, in combination with ribose
and pyrophosphate bond modifications, have successfully brought the development of
effective inhibitors of DHs.50, 51
Some of the DH inhibitors have been found to inhibit
also NADK, responsible for the phosphorylation of the 2′-OH of β-NAD+ to generate β-
NADP+.52
As mentioned in 1.2.1, NaM, the reaction product of ARTs, PARPs, sirtuins, ADPRCs
and NADases activities, is a common inhibitor for these enzymes, in particular as a
competitive inhibitor for ARTs and PARPs,53, 54
a non-competitive inhibitor for
sirtuins,55
and a mixed inhibitor for ADPRCs56
and NADases.24
Many nicotinamide
derivatives have been developed as potential inhibitors for these enzymes (e.g. 8-13)
(Figure 1.14). All the developed nicotinamide derivatives contain in their structure
essential features to act as inhibitors of ARTs, PARPs and sirtuins. For instance, a
primary or secondary amide, responsible for the interactions between NaM and the
enzyme active sites, is engaged in H-bonding by its carbonyl and amine groups,31
and
its anti conformation is specifically required for the compounds activity toward ARTs
and PARPs. In addition, an aromatic ring more electron-rich than the pyridine, and a
non cleavable bond contribute to the inhibitory activity of nicotinamide derivatives.53, 54
β-NADH (14) and β-NAAD+ (15), where NaM is substituted with the reduced pyridine
Chapter 1
27
ring and NA respectively, act as inhibitors of sirtuins.57
Moreover, BAD (1), AAD (16)
and HHD (17) inhibit ADPRCs activity (Figure 1.14).58
All these inhibitors give an
example of the essential nature of the nicotinamide portion for the activity of NAD+-
dependent enzymes.
Figure 1.14 Inhibitors of DHs, NADKs, ARTs, PARPs, sirtuins, ADPRCs and NADases based on
nicotinamide modifications. Abbreviations used: NAD+, nicotinamide adenine dinucleotide; BAD,
benzamide adenine dinucleotide; C-NAD, C-nicotinamide adenine dinucleotide; C-PAD, C-picolinamide
adenine dinucleotide; NADH, nicotinamide adenine dinucleotide (reduced form); NAAD+, nicotinic acid
adenine dinucleotide; AAD, bis-adenine dinucleotide; HHD, bis-hypoxanthine dinucleotide.
Acceptor H-bondingDonor H-bonding
Not cleavable bond
Chapter 1
28
1.3.1.2 Nicotinamide ribose modification
The nicotinamide ribose plays a fundamental role in the activities of NAD+-dependent
enzymes. In particular, as shown in 1.2.1, it is actively involved in the reaction
mechanisms of NAD+-consuming enzymes with a common NADase activity due to the
formation of a stabilized oxocarbenium intermediate. Its modification has been
undertaken to generate inhibitors for NAD+-consuming enzymes, such as ARTs,
PARPs, sirtuins, ADPRCs and NADases. However, it has also been targeted to improve
the efficacy of inhibitors for DHs50
and NADKs.52
The most common modifications of
the nicotinamide ribose have been suggested by the knowledge of the reaction
mechanism for enzymes with NADase activity (Figure 1.9). The substitution of O-1 of
the ribose with a methylene group (CH2) causes the loss of stabilization of the
oxocarbenium intermediate by the oxygen lone pairs; therefore, the replacement of the
ribose ring with a 2,3-dihydroxycyclopentane ring generates a dinucleotide
diastereomers mixture, containing carba-NAD+ (1, with D-2,3-dihydroxycyclopentane
ring) and pseudo carba-NAD+ (2, ψ carba-NAD
+, with L-2,3-dihydroxycyclopentane
ring) (Figure 1.15).59
In particular, ψ carba-NAD+ is more active than carba-NAD
+ as an
inhibitor of ARTs, PARPs, sirtuins, NADases and CD38,54, 60
while both derivatives are
inactive toward ADPRC from Aplysia californica,60
shown to be insensitive to ribose
modifications.58
By contrast to its inhibitory properties toward enzymes with NADase
activities, carba-NAD+ acts as a NNS for DHs, while ψ carba-NAD
+ is completely
inactive.59
The substitution of the 2′′-OH of the nicotinamide ribose with a fluorine (F) deletes its
H-bonding interactions with glutamate residues or electron-withdrawing groups in the
active site, which help in stabilizing the oxocarbenium intermediate. Further
modifications, such as the change of the ribose with arabinose, where the 2′′-OH (F)
points toward the other face of the ribose; or the removal of the 2′′-OH in 2′′-deoxy
nicotinamide ribonucleoside derivatives,61
achieve the same destabilizing effect on the
oxocarbenium intermediate. Therefore, 2′′-deoxy nicotinamide ribonucleoside
derivatives (3-5), 2′′-FNAD+ (6), ara-NAD
+ (7), ara-2′′-FNAD
+ (8) and ara-2′′-FNMN
(9) are all good inhibitors of NAD+-dependent enzymes (Figure 1.15), in particular of
DHs, in combination with nicotinamide modifications,50
NADKs,52
NADases,62
and
Chapter 1
29
CD38.30, 61, 63
Finally, the change of nicotinamide ribose conformation from β-D-ribose
to α-D-ribose generates α-NAD+ that inhibits sirtuins.
57
Figure 1.15 Inhibitors of DHs, NADKs, sirtuins and NADases based on nicotinamide ribose
modifications. Abbreviations used: NAD+, nicotinamide adenine dinucleotide; 2′′-FNAD
+, 2′′-
fluoroNAD+; ara-NAD
+, arabinose-NAD
+; ara-2′′-FNAD
+, arabinose-2′′-fluoroNAD
+; ara-2′′-FNMN,
arabinose-2′′-fluoro nicotinamide mononucleotide.
1.3.1.3 Pyrophosphate bond modification
The β-NAD+ structure is largely dynamic because of the variety of flexible bonds and
rotational angles that it can assume in generating different conformations.64
The
pyrophosphate bond that joins the two nucleosides is the most flexible part of the
molecule because of the three dihedral angles, ζ, α and γ, able to rotate and confer
different orientations and conformations to the nucleosides (Figure 1.16). Due to its
high flexibility, the pyrophosphate bond is generally not strictly involved in recognition
by enzymes. However, its modification in the synthesis of dinucleotide derivatives is
very important because it confers stability to potential inhibitors or NNSs against in vivo
enzymatic degradation by nucleases, which are often responsible for the cleavage of
Chapter 1
30
nucleic acid phosphodiester or dinucleotide pyrophosphate bonds. Therefore,
modifications of the pyrophosphate bond have been used to inhibit nucleases, or, in the
particular case of β-NAD+, NPPs.
65 However, its coupling to other modifications has
allowed generation of effective DHs, NADKs and NAD+-consuming enzyme inhibitors.
The most common modifications on the pyrophosphate bond (Figure 1.16) involve the
substitution of the oxygen bridge between the two phosphates with a CH2 or
difluoromethylene (CF2) group to generate phosphonate derivatives (1, 2).50, 66
These
modifications do not alter the three-dimensional shape of the molecule since the size of
these groups is similar to the oxygen, but they do provide resistance to attack by
nucleases and make the dinucleotides more lipophilic and able to penetrate inside cells.
Moreover, CF2 is a better bioisostere than CH2, since the electron-withdrawing effect of
the fluorine atoms increases the pKa of the phosphate groups, making them more similar
to the natural pyrophosphate bond. Other substitutions67
can be the introduction of
phosphatephosphonate groups (3, 4), sulphonamide groups (5), amidephosphonate
groups (6), thioethylenephosphonate (7) or disulphide68
(8) groups.
Figure 1.16 Inhibitors of DHs, NADKs and NPPs based on pyrophosphate bond modifications.
Abbreviations used: NAD+, nicotinamide adenine dinucleotide.
Chapter 1
31
1.3.1.4 Adenine modification
The adenine moiety in β-NAD+ is fundamental for recognition and binding by enzymes.
With the exception of NAD+-dependent DNA ligases, sirtuins and ADPRCs, DHs,
NADKs and all other NAD+-consuming enzymes do not actively involve the adenine
moiety in their catalytic mechanisms. Therefore, its modification within the β-NAD+
structure is generally aimed at the synthesis of NNSs for these enzymes. As discussed in
1.3, the NNS activity implies a competitive inhibition with β-NAD+. Hence, from the
design of NNSs with high affinity for the enzyme (low Km) and slow turnover (low kcat),
it is possible to generate effective inhibitors. The latter is a common case and a good
example is given by NNSs for ADPRC from Aplysia californica. This enzyme has
shown broad substrate tolerance related to the adenine moiety, which has been
substituted with guanine (1), hypoxanthine (2), xanthine and ε-adenine to generate
NNSs which are also good inhibitors for ADPRC (Figure 1.17).58, 69
Whilst the NNS activity often coincides with the competitive inhibition of the NS, the
reverse is not always true. For instance, 8-aryl/heteroaryl NAD+ derivatives are
inhibitors but not NNSs for sirtuins (3, Figure 1.17),70
indicating that the substitutions in
position 8 on the adenine ring allow the binding of the compounds to the enzyme active
site but inhibit their metabolism (see 5.2.2). Another example is given by 2-fluoroNAD+
(2-FNAD+) (4, Figure 1.17);
71 indeed, this compound is cleaved by ADPRC from
Aplysia californica on the nicotinamide N-glycosidic bond, but its cyclization is
inhibited by the fluorine in position 2 on the adenine ring, with consequent release of 2-
FADPR. Therefore, the compound acts as a substrate for the NADase activity of
ADPRC, but as an inhibitor toward its main cyclase activity.
Beyond the adenine-modified NAD+ derivatives, adenine-based compounds (5) and
adenine-modified adenosine derivatives (6, 7) (Figure 1.17) have been developed,
respectively, as inhibitors of sirtuins72
and NAD+-dependent DNA ligases,
73, 74
indicating the strong influence of the adenine moiety in both enzymes.
Chapter 1
32
Figure 1.17 Inhibitors of ADPRCs, NPPs, sirtuins and DNA ligases based on adenine modifications.
Abbreviations used: NAD+, nicotinamide adenine dinucleotide; NGD
+, nicotinamide guanine
dinucleotide; NHD+, nicotinamide hypoxanthine dinucleotide; 2-FNAD
+, 2-fluoroNAD
+.
1.3.2 NAD+ derivatives as non-natural substrates of NAD+-dependent enzymes
The development of NNSs can help to investigate the active sites and catalytic
mechanisms of enzymes. Therefore, targeted modifications and labelling are required to
generate useful biochemical tools. As mentioned in 1.3.1.4, the adenine modifications
are generally aimed at the synthesis of NNSs for NAD+-dependent enzymes. The
adenine base has been modified preferentially in a few favourite positions, such as N-1,
NH2 in 6 position and C-8 position. The correlation between different NNS activities
and the modifications in different positions highlight the structure and the fundamental
interactions occurring between the β-NAD+ scaffold and the active sites of enzymes.
However, modifications on the nicotinamide ring can also generate NNSs for NAD+-
dependent enzymes, as long as they keep unchanged the labile N-glycosidic bond or the
redox activity of the nicotinamide ring that is still used as a substrate.
Chapter 1
33
NNSs for DHs have mostly been modified in N-1 (4-6), NH2 in 6 position (1, 2, 7-9) or
both (3) on the adenine ring (Figure 1.18).75, 76
Other NAD+ derivatives have been
synthesized substituting the amide group on the pyridine ring (14-18), still maintaining
its redox activity (Figure 1.18).77, 78
Different DHs have shown selective specificities
toward the differently substituted NAD+ derivatives, demonstrating the existence of
structural differences within the same class of enzymes.
Figure 1.18 NAD+ derivatives with NNS activities toward NAD
+-dependent enzymes. Abbreviations
used: NAD+, nicotinamide adenine dinucleotide; NHD
+, nicotinamide hypoxanthine dinucleotide; NFD
+,
nicotinamide formycin dinucleotide; ε-NAD+, 1,N
6-etheno-NAD
+; NGD
+, nicotinamide guanine
dinucleotide; NXD+, nicotinamide xanthine dinucleotide; 3AcPAD
+, 3-acetyl pyridine adenine
dinucleotide; 3AlPAD+, 3-aldehyde pyridine adenine dinucleotide; thio-NAD
+, thio-nicotinamide adenine
dinucleotide; ADPR-pNP, adenosine diphosphate ribose para-nitrophenol.
NNSs for NAD+-consuming enzymes have mostly been modified on the adenine ring
(1, 3, 10-13, Figure 1.18). Also in this case, each enzyme has shown preferences toward
particular substrates depending on their active site. For example, PARPs has shown
preferences toward 12 compared to 1 or 3,79
indicating the importance of a free amino
group on the adenine ring; and, furthermore, PARPs, sirtuins and NADases have higher
affinity for 10 compared to 11.14, 80, 81
In addition, some of the pyridine-modified
derivatives have been used as NNSs for sirtuins (14, 18, 19),57
and PARPs (21).82
Many
Chapter 1
34
substitutions introduced in the NNSs shown in Figure 1.18 have allowed the
development of binding and kinetic assays for NAD+-dependent enzymes, as discussed
in the next section.
At this point it is worth to introduce the most common NNS of β-NAD+, 1,N
6-etheno-
NAD+ (ε-NAD
+, Figure 1.19), which has been used since the 1970s to investigate many
NAD+-dependent enzymes.
83 Its structure is characterized by the introduction of an
etheno bridge between the N-1 and the exocyclic amino group in 6 position on the
adenine ring; this modification, extending the electron delocalization along the adenine
ring, makes the latter strongly fluorescent. Owing to this property, ε-NAD+ has been
extensively used as a fluorescent probe for NAD+-dependent enzymes; indeed, the small
modification on the adenine ring does not generally interfere with binding to enzyme
active sites, allowing structural studies of the relevant enzymes. Moreover, changes in
fluorescence following the enzymatic metabolism of the compound by the action of
NAD+-consuming enzymes have allowed the development of fluorescence-based assays
for accurate kinetic and mechanistic studies, which will be examined later in this
chapter; the characteristic fluorescence properties of this compound and its derivatives
will also be considered further in Chapter 3 (see 3.7.1).
Figure 1.19 Chemical structure of ε-NAD+.
Despite its extensive use for the study of many NAD+-dependent enzymes, ε-NAD
+ has
shown some limitations due to its modification on the adenine ring. Indeed, as will be
shown later, ε-NAD+ is not a good substrate for PARPs (see 1.4.2), ADPRC from
Aplysia kurodai56
and some NPPs (see 4.1.2). Moreover, its use for the investigation of
Chapter 1
35
sirtuins and NAD+-dependent DNA ligases has never been reported, indicating its
unsuitability to act as an inhibitor and/or NNS toward these enzymes. Hence, the
development of novel classes of NAD+ derivatives with different features compared to
ε-NAD+ would provide a better insight into the knowledge of many NAD
+-dependent
enzymes.
1.4 Assays for NAD+-consuming enzymes
The investigation of enzyme structures, reaction mechanisms and kinetic measurements
is fundamental to the understanding of the enzyme activities and how they correlate to
major biological processes. Therefore, the development of NNSs as biochemical probes
allows us to answer several biochemical questions. As previously discussed, the
chemical modification of different positions on adenine or pyridine rings, and how they
influence the corresponding NNS activities, provides an improved understanding of the
active site structure and the requirement of fundamental moieties in the natural molecule
scaffold. Moreover, the labelling of NNSs with groups that are detectable by sensitive
techniques allows the development of more sensitive binding, kinetic and screening
assays, compared to those using the NS. These assays generally exploit the labelled
NNSs or the corresponding reaction products to monitor their consumption/formation in
a continuous or non-continuous format by sensitive techniques. Alternatively, the
labelling of the remaining β-NAD+ or reaction products can be performed in a second
step of the assay to provide steady-state/endpoint measurements of the enzymatic
activities.
Herein, attention will be focused on the existing assays and their principles to monitor
the activities of NAD+-consuming enzymes. As will be shown, assays in a continuous
format have been generally developed for those enzymes having β-NAD+ as the only
substrate. In contrast, steady-state/endpoint assays have been the only way to monitor
the activity of NAD+-consuming enzymes that have more than one substrate involved in
their reaction. Indeed, the different complexities of the reaction mechanisms involving
one or more substrates reflects on the substrate/co-substrate binding specificity, which
becomes higher if the enzyme has to coordinate the entrance, binding in the active site
and metabolism of more than one substrate. Therefore, the development of labelled
Chapter 1
36
NNSs which are tolerated by enzymes with complex reaction mechanisms is an open
challenge. In this perspective, novel classes of NAD+ derivatives with NNS activities
toward NAD+-consuming enzymes could be exploited for the development of new
advantageous assays.
1.4.1 Cyanide and dehydrogenase assays
One of the oldest methods to detect NAD+-consuming enzyme activities is the reaction
of the quaternary nitrogen of NaM in β-NAD+ with cyanide to form a complex that
absorbs at 325 nm (Figure 1.20).84
Adenine nitrogens, free NaM, or
dihydronicotinamide in β-NADH do not interfere with this reaction, which is specific
for quaternary amines. Moreover, β-NMN can react with cyanide and give a complex
with the same spectral properties of β-NAD+, thus interfering with the assay. While the
activity of NADases can be monitored by an absorbance decrease at 325 nm,
proportional to the consumption of β-NAD+, the activity of NPPs cannot be detected,
since β-NAD+ and NPP‟s reaction product, β-NMN, give exactly the same absorbance
reading.
A parallel method to detect changes in β-NAD+ concentration consists of the reaction of
the remaining β-NAD+ with DHs to generate β-NADH, formation of which can be
monitored by absorbance at 340 nm or emission at 475 nm (Figure 1.20),84
where β-
NAD+ does not interfere. Indeed, β-NAD
+ and β-NADH are both characterized by a
maximum absorbance at 260 nm with an extinction coefficient of 16900 M-1
·cm-1
due to
the adenine ring; in addition, β-NADH has another absorbance peak at 340 nm with an
extinction coefficient of 6220 M-1
·cm-1
, due exclusively to the reduced nicotinamide
ring. Exciting β-NAD+ and β-NADH at 360 nm, the compounds emit respectively at
445 and 475 nm; however, β-NAD+ emission is very weak compared to that of β-
NADH and, therefore, working with in vivo concentrations is possible to
fluorimetrically detect only β-NADH.85, 86
Identical values for the activity of NADases are obtained by the measurements via both
cyanide and DH methods, unless β-NAD+ contains β-NMN impurities that give negative
interference. A combination of cyanide and DH methods allows determination of
Chapter 1
37
whether NPPs, NADases or both activities are responsible for the β-NAD+
consumption. Indeed, a positive absorbance difference between the readings obtained
with cyanide and DH methods indicates the presence of NPPs activity beyond that of
the NADases.
Figure 1.20 Cyanide and dehydrogenase (DH) assays to detect NAD+-consuming activities. Abbreviations
used: NAD+, nicotinamide adenine dinucleotide; NMN, nicotinamide mononucleotide; ADP, adenosine
diphosphate; Pi, phosphate; NAD-CN, NAD-cyanide complex; NADH, reduced NAD+.
1.4.2 -NAD+ and its application for NAD+-consuming enzyme assays
ε-NAD+ (Figure 1.21), as mentioned in 1.3.2 and discussed in detail in Chapter 3, has
been largely employed as a fluorescent NNS to monitor the activity of NAD+-
consuming enzymes. Its fluorescence properties, more than its absorbance properties,
have been used for the development of continuous or steady-state fluorescence-based
assays depending on the examined enzymes, or to gain more sensitivity in HPLC
analysis. As will be discussed in more detail in Chapter 3, ε-NAD+ is characterized by a
strong internal fluorescence quenching due to stacking interactions between the ε-
adenine and the nicotinamide rings in the intact β-NAD+ structure.
83, 87, 88 When the ε-
NAD+ structure is disrupted by the activities of NAD
+-consuming enzymes, the
fluorescence of the ε-adenine moiety is re-established with a consequent fluorescence
Chapter 1
38
increase, which can be spectroscopically monitored. NPPs,89
ADPRCs,69, 90
NADases87,
91 and ARTs
92, 93 activities can be evaluated by ε-NAD
+ continuous fluorescence-based
assays. Indeed, all enzymatic products from the activity of these enzymes are more
fluorescent than ε-NAD+ because of the disruption of the internal quenching between
the aromatic rings (Figure 1.21). Therefore, a fluorescence increase during the
enzymatic reactions can be detected at 410 nm (λex = 300 nm), allowing monitoring of
enzyme activities and kinetic measurements.
Figure 1.21 ε-NAD+ as a NNS of NADases, NPPs, ADPRCs and ARTs for the development of
continuous fluorescence-based assays. A fluorescence increase is monitored at 410 nm (λex = 300 nm)
during the enzymatic reactions of the above NAD+-consuming enzymes. Abbreviations used: ε-NAD
+,
1,N6-etheno-nicotinamide adenine dinucleotide; ε-ADPR, 1,N
6-etheno-adenosine diphosphate ribose; ε-
AMP, 1,N6-etheno-adenosine monophosphate; cεADPR, cyclic 1,N
6-etheno-adenosine diphosphate
ribose.
Chapter 1
39
Unfortunately, because all these enzymes use ε-NAD+ as a NNS, it is not possible
simply by fluorescence measurements to distinguish the single enzymatic activities
when they are all present. In addition, since these enzymes belong to the same class of
ecto-enzymes, this causes a lack of enzyme compartment differentiation, not allowing
any enzymatic activity distinction in vitro/in vivo. Consequently, only the discovery of
fluorescent NNSs selective for particular enzymes or HPLC analysis can allow the
distinction of the different enzymatic activities in vitro/in vivo. Hence, modifications to
the ε-NAD+ structure have been performed in the attempt to generate specific substrates
for different enzymes. For example, ε-PdAD+ and ε-hy
4PdAD
+ (Figure 1.22) have
similar fluorescent properties to ε-NAD+
, with ε-PdAD+ acting as a specific substrate
toward NPPs, but not toward NADases.89
Moreover, the fluorescent properties of ε-
NAD+ and its different enzymatic products have allowed the development of HPLC
assays with fluorescence detection, which are much more sensitive than those using the
natural β-NAD+.92, 94
Figure 1.22 ε-NAD+ and its derivatives. Abbreviations used: ε-NAD
+, 1,N
6-etheno-nicotinamide adenine
dinucleotide; ε-PdAD+, 1,N
6-etheno-pyridine adenine dinucleotide; ε-hy
4PdAD
+, 1,N
6-etheno-4-
hydrazinocarbonyl-pyridine adenine dinucleotide.
PARPs use ε-NAD+ as a NNS, but they do not have a very good affinity for it since the
formation of poly(ε-ADPR) is very slow and the length of the polymers much shorter
compared to those from β-NAD+.79
In principle, a fluorescence increase should be
observed in relation to the activity of PARPs that catalyzes the removal of NaM and,
Chapter 1
40
therefore, the disruption of the internal quenching in ε-NAD+. However, a strong
fluorescence quenching occurs in the polymers of ε-ADPR, not allowing a continuous
monitoring of the activity of PARPs by increase in fluorescence emission.95
By contrast,
the degradation of poly(ε-ADPR) generates a fluorescence increase due to the release of
ε-ADPR units, which gives information about the extent of PARPs activity.79
The
poly(ε-ADPR) degradation can be performed by phosphodiesterases, but not by
NADases or PARP-glycohydrolases (PARGs);79, 96
the latter enzymes are indeed
inhibited by poly(ε-ADPR).96
To avoid the slow formation of poly(ε-ADPR) using ε-
NAD+ as a NNS for PARPs, poly(ε-ADPR) can be generated in situ from poly(ADPR),
obtained by the reaction of PARPs with the natural β-NAD+, by staining with
chloroacetaldehyde.79, 96
Alternatively, poly(ADPR) can be degraded by NPPs to ADPR
units, which can be stained in situ with chloroacetaldehyde to obtain ε-ADPR.97
Since
the activity of PARPs cannot be directly monitored by fluorescence changes using ε-
NAD+ as a NNS, an immunohistochemical reaction between the ε-adenosine moiety and
a specific antibody has been developed to detect in situ the formation of poly(ε-
ADPR);95
this method is more sensitive and specific than another immunohistochemical
reaction involving the natural adenosine moiety and allowing the in situ detection of
poly(ADPR).98
1.4.3 NGD+, NHD+ and NXD+ for ADPRC fluorescence-based assays
NGD+, NHD
+ and NXD
+ have been used for the development of continuous
fluorescence-based assays to monitor the activity of ADPRCs.69
Indeed, these
compounds, not fluorescent in the form of dinucleotides, become strongly fluorescent
after their cyclization by ADPRCs (Figure 1.23).
The acquired fluorescence properties seem to be due to the cyclization of the
compounds in N-7, generating a different electron delocalization along the ring because
of the generated positive charge. Therefore, a fluorescence increase during the
enzymatic reaction can be detected at 410 nm (λex = 300 nm), allowing monitoring of
the activity of ADPRCs and estimation of kinetic inhibition parameters for this enzyme.
Detailed discussions about these NNSs for ADPRC from Aplysia californica and their
assays will be provided in Chapter 4 (see 4.2.3).
Chapter 1
41
Figure 1.23 NGD+, NHD
+ and NXD
+ as NNSs of ADPRCs for the development of continuous
fluorescence-based assays. A fluorescence increase is monitored at 410 nm (λex = 300 nm) during the
enzymatic reaction of ADPRCs with NGD+, NHD
+ and NXD
+. Abbreviations used: β-NMN,
nicotinamide mononucleotide; NGD+, nicotinamide guanine dinucleotide; NHD
+, nicotinamide
hypoxanthine dinucleotide; NXD+, nicotinamide xanthine dinucleotide; cGDPR, cyclic guanosine
diphosphate ribose; cIDPR, cyclic inosine diphosphate ribose; cXDPR, cyclic xanthosine diphosphate
ribose.
1.4.4 6 and 8-alkyne NAD+ for NADase activity assays
As discussed in 1.2.1, NADases, ARTs, PARPs and sirtuins are characterized by
NADase activity since they all cleave the nicotinamide N-glycosidic bond with release
of NaM. Then, the generated oxocarbenium intermediate can react in different ways
depending on the considered enzyme.
A new assay has been developed by Lin and co-workers involving the use of 6 or 8-
alkyne NAD+ as NNSs to label and monitor the activities and the targets of NADase
(CD38), ARTs, PARPs and sirtuins.14, 80, 81
The principle of the assay is based on the
labelling of the different ADPR enzymatic products by reaction of the alkyne moiety,
introduced in the 6 or 8 position on the adenine ring, with rhodamine or biotin tagged
Chapter 1
42
azides via click chemistry (Figure 1.24). Generally, 6-alkyne NAD+ has been shown to
be a better substrate than 8-alkyne NAD+ for the enzymes examined.
In the case of the activities of ARTs and PARPs, the labelling of mono-ADPR or poly-
ADPR acceptors via rhodamine has allowed their detection in gel electrophoresis by
fluorescence; by contrast, the labelling via biotin has allowed their isolation and
purification by affinity chromatography, with identification by LC-MS/MS.80
In the
case of sirtuins, the assay has been used to monitor their suspected ART activity.
However, even if 6 or 8-alkyne NAD+ have been shown to be NNSs for the
deacetylation activity, they were not able to detect any ART activity exerted by sirtuins,
probably because of its weakness compared to the deacetylation process.14
Finally, the
alternative substrates have been used to covalently label CD38.81
Indeed, it is known
that ara-2′′-FNAD+ inhibits CD38 by covalently binding the ADPR moiety by C-1′′ to a
glutamate residue.63
The introduction of a 6 or 8-alkyne moiety on ara-2′′-FNAD+ has
therefore allowed fluorescent labelling of CD38 in living cells.
Figure 1.24 6 and 8-alkyne NAD+ as NNSs of NADases, ARTs, PARPs and sirtuins for the development
of fluorescence or immunohistochemical assays. After the enzymatic reaction using 6 or 8-alkyne NAD+
as NNSs, the different enzymatic products can undergo click chemistry to introduce the biotin or the
rhodamine tag.
Chapter 1
43
1.4.5 β-NAD+ quantification by reaction with acetophenone
Another non-continuous assay to monitor the activity of PARPs and sirtuins is based on
the quantification of the unreacted β-NAD+ by its reaction with acetophenone to
generate a fluorescent NAD+ derivative (1, Figure 1.25).
99-101 The treatment of the
enzymatic reactions with solutions of KOH and acetophenone, followed by addition of
formic acid and heating, leads to the formation of a cyclized fluorescent product, 1,
which can be detected by fluorescence emission at 444 nm (λex = 372 nm); at this
wavelength, β-NAD+ and NaM do not have any intrinsic fluorescence, therefore they
cannot interfere with the assay. Even though it is an endpoint assay requiring different
reaction steps, the possibility to perform it on 96 well-plates is a great advantage in
terms of large screening of inhibitors.
Figure 1.25 Quantification of the unreacted β-NAD+ by its reaction with acetophenone to generate the
fluorescent NAD+ derivative, 1. A KOH solution is added to the enzymatic reaction to generate the a-
carbinol intermediate. Then, the addition of acetophenone generates a highly reactive a-adduct, which
spontaneously cyclizes to a cyclized non-fluorescent a-adduct; the latter, after heating at 110 °C,
dehydrogenates to give the fluorescent NAD+ derivative, 1.
1.4.6 A continuous colourimetric assay for the detection of PARP activity
A continuous colourimetric assay has been developed for the study of the activity of
PARPs; its principle is based on the colour changes occurring during the enzymatic
reaction of an NAD+ derivative, used as a NNS of PARPs, in which the nicotinamide
ring is substituted with a para-nitrophenol ring.82
The latter is a well known pH
indicator, indeed, it is colourless at acidic/neutral pH and bright yellow at basic pH (pKa
= 7.16), when it is in the form of para-nitrophenolate. PARPs are able to cleave the O-
glycosidic bond between the C-1′′ of the ribose and the para-nitrophenol ring, which is
released under its phenolate form to be monitored by absorbance increase at 405 nm
(Figure 1.26).
Chapter 1
44
Figure 1.26 ADPR-para-nitrophenol (ADPR-pNP) as a NNS of PARPs for the development of a
continuous colourimetric assay. During the enzymatic reaction, the release of the yellow para-
nitrophenolate can be monitored at 405 nm.
1.4.7 Alternative assays for the detection of sirtuins activity
Many assays have been developed in the past years to test the activity of sirtuins. Most
of the initial assays involved the use of radioactive β-NAD+, labelled with
14C/
32P on the
nicotinamide ring or the pyrophosphate bond, or, alternatively, radioactive acetylated
peptides, labelled with 3H on the acetate group.
102 However, radioactive measurements,
despite their sensitivity, involve a considerable health hazard, cost and waste issues.
More recently, several non-continuous fluorescence-based assays to monitor the activity
of sirtuins have been developed: one of them, described in 1.4.5, is based on the
detection of the unreacted β-NAD+ by transformation of the latter in a fluorescent
derivative. Other assays rely on fluorescence changes during the enzymatic reaction of
internally quenched fluorescent peptide substrates. A common assay based on this
principle is the Fluor de Lys, commercially available in kits, which uses as a substrate a
peptide comprising amino acids 379-382 of human p53 (Arg-His-Lys-Lys[Ac]),
containing at the C terminus the fluorophore AMC (7-amino-4-methylcoumarin). Other
assays are based on the same principle, using as substrates MAL (Boc-Lys[Ac]-
AMC)103
or Z-MAL (Cbz-Lys[Ac]-AMC).104
In the first step of the assay, sirtuins are
incubated with β-NAD+ and the fluorescent quenched acetylated peptide substrate (for
instance, Z-MAL, Figure 1.27) in the reaction buffer. The peptide, deacetylated by the
activity of sirtuins, ZML, is a trypsin substrate; therefore, after the addition into the
reaction mixture of a stop solution containing trypsin, the fluorescent product, AMC is
released in solution with consequent fluorescence increase. The developed fluorescence
is proportional to the deacetylation performed by sirtuins.103, 104
This assay has been
Chapter 1
45
used for the biological evaluation of 8-aryl/heteroaryl substituted NAD+ derivatives
described in Chapter 5.
Figure 1.27 Fluorescence-based assay for sirtuins activity. Z-MAL (Cbz-Lys[Ac]-AMC) is incubated with
sirtuins and β-NAD+ to give the deacetylated product, ZML (Cbz-Lys-AMC). The latter is hydrolysed by
trypsin to release the fluorophore, AMC (7-amino-4-methylcoumarin), and Z-lysine. The fluorescence is
proportional to the deacetylation exerted by sirtuins.
1.4.8 Assays for the detection of DNA ligase activity
The gel electrophoresis technique is commonly used to monitor the DNA ligation
experiments. Its principle is based on the detection of fluorescently or radioactively
labelled single strand nicked DNA and its corresponding eventually joined DNA,
migrating differently on the gel because of their different molecular sizes. This
technique, exploiting a fluorescently labelled single strand nicked DNA, has been used
to evaluate the NAD+ derivatives synthesized in Chapter 2 (more details in Chapter 5,
see 5.1.1) toward the ligation activity of EcLigA. However, the gel electrophoresis
technique is generally time-consuming and laborious, especially in the screening of
many compounds, and its results strongly depend on the batches of enzyme activity and
DNA substrates. For this reason, other assays for high-throughput screening of potential
inhibitors of DNA ligases have been developed to simplify and overcome the
Chapter 1
46
limitations of the gel electrophoresis assay. Some of them are based on FRET
(fluorescence resonance energy transfer) signals generated upon ligation of DNA
strands carrying a fluorescence donor and acceptor (Figure 1.28 a);73
others on the
fluorescence changes upon ligation of molecular beacons carrying a fluorophore (Figure
1.28 b),105
or a fluorophore and a quencher at the extremities.106
These assays, although
non-continuous and involving different steps (e.g. ligation reaction, denaturation,
washing), allow high-throughput screening of potential inhibitors of DNA ligases in 96
well-plates with great accuracy and reproducibility.
Figure 1.28 Ligation assays based on the generation of FRET signals (a), or on the fluorescence detection
of ligated molecular beacons105
(b) upon ligation and denaturation conditions.
1.5 Project Objectives
Interest towards NAD+-consuming enzymes has increased in the past years since a
deeper knowledge of fundamental biological processes has shown the involvement of
this class of enzymes in their regulation and, consequently, in a number of human
diseases. Therefore, the necessity to further investigate their structure and catalytic
mechanism and, eventually, to modulate or block their activities has brought an
admirable scientific effort in the development of efficient and accurate assays to
monitor their enzymatic reactions and screen large numbers of inhibitors.
As reviewed within this chapter, many improvements have been made in targeting and
regulating these enzymes. Since β-NAD+ is the common substrate of NAD
+-consuming
enzymes, a major effort has been made in generating NAD+ derivatives that could act as
FAM5′
TAMRAP
3′
3′
OH
5′
FAM5′
TAMRA3′
3′
5′
FRET
32-mer
18-mer 11-mer
32-mer
29-mer
Ligationdenaturation
a b
Chapter 1
47
inhibitors and/or NNSs for these enzymes. A continuous increase and updating of
information about the reaction mechanism and active site structure of NAD+-consuming
enzymes has allowed more specific design of compounds that could show some
biological activity toward them.
At the outset of this project, a review of the existing NAD+ derivatives and their
activities toward NAD+-consuming enzymes has shown that, among the different
modifications on the β-NAD+ scaffold, those on the adenine ring can generate different
biological activities depending on the enzymes examined. For example, the adenine
modifications are tolerated by a large number of NAD+-consuming enzymes, such as
NADases, ADPRCs, ARTs and PARPs, generating NNSs that have been exploited for
the development of binding and kinetic assays. On the other hand, two interesting
classes of enzymes, such as NAD+-dependent DNA ligases and sirtuins, belonging to
the NAD+-consuming enzymes but characterized by complex reaction mechanisms, had
shown a strong involvement of the adenine moiety in the enzyme binding and reaction
mechanism, suggesting that its modifications can generate potential inhibitors
interfering with their activities.
This project has been developed on the basis of these arguments, in an attempt to
generate novel adenine-modified NAD+ derivatives, that could provide new biochemical
information about NAD+-consuming enzymes. The main objectives proposed in this
project are summarised below and in Figure 1.29.
Synthesis of novel NAD+ derivatives, modified in the position 2, 6 or 8 of the
adenine ring with aryl/heteroaryl groups by palladium-mediated cross-couplings
(Figure 1.29 a) (see Chapter 2). The successful application of these couplings would
offer the opportunity to insert many different chemical groups on the adenine ring,
keeping intact its skeleton. These modifications, although already applied to the
synthesis of nucleosides, nucleotides and DNA derivatives, have never been used
for the synthesis of NAD+ derivatives.
Chapter 1
48
Spectroscopic analysis of the novel NAD+ derivatives in view of their potential
activity as NNSs for NAD+-consuming enzymes and, therefore, their application to
the development of sensitive assays based on spectroscopic techniques (see Chapter
3).
Evaluation of the novel NAD+ derivatives toward NAD
+-consuming enzymes as
inhibitors for drug discovery and/or NNSs for the development of advantageous,
continuous or non-continuous, assays as an alternative to those already existing
(Figure 1.29 b). NPPs, NADases and ADPRCs were chosen as model enzymes to
test the activities of the NAD+ derivatives because of the simplicity of their reaction
mechanism, involving β-NAD+ as the only substrate (see Chapter 4). In contrast,
NAD+-dependent DNA ligases and sirtuins were chosen as examples of complex
NAD+-consuming enzymes, using β-NAD
+ as a co-substrate (see Chapter 5).
Figure 1.29 Synthetic (a) and biological (b) targets of the project. The synthetic targets (a) are
represented by the novel NAD+ derivatives, modified in the position 2, 6 or 8 of the adenine ring with
aryl/heteroaryl substituents (R). The biological targets (b) are represented by the listed NAD+-consuming
enzymes, divided in two classes depending on the activity of β-NAD+ as a substrate/co-substrate in their
reactions.
β-NAD+ NAD+-consuming enzyme
substrate
NPP
NADase
ADPRC
co-substrate
NAD+-dependent DNA ligase
sirtuins
a
b
Chapter 1
49
Analysis of the different activities exerted on different enzymes by NAD+
derivatives substituted in the same position on the adenine ring and, therefore, their
ability to target specific biological processes. The enzyme specificity is a
fundamental requirement in the drug development for in vitro/in vivo studies (see
Chapter 6).
Therefore, the development of novel adenine-modified NAD+ derivatives as
biochemical probes for NAD+-consuming enzymes provides new tools for the
investigation of this important class of enzymes.
Chapter 2
50
2 SYNTHESIS OF ARYL/HETEROARYL ADENINE-MODIFIED NAD+ DERIVATIVES
2.1 Synthetic strategies to aryl/heteroaryl adenine-modified NAD+ derivatives
β-NAD+ is a fundamental biological molecule found in all living cells and used by a
variety of enzymes involved in many different biological processes. Its structure with
two labile high-energy bonds, i.e. the pyrophosphate and the nicotinamide N-glycosidic
bond (see 1.1), underpins the biological activity of β-NAD+ as a substrate/co-substrate
of NAD+-consuming enzymes, but also makes it an extremely fragile molecule.
5, 7, 13
Indeed, β-NAD+ can be subject to degradation even under mild conditions, which
complicates its chemical synthesis and derivatisation. Nevertheless, given the
importance of β-NAD+ in biology, the development of NAD
+ derivatives is of
considerable interest and importance, as such derivatives can be used as biochemical
tools to investigate NAD+-dependent enzymes, their reaction mechanisms and their
biological roles.
Owing to its limited chemical stability, the chemical modification of β-NAD+ is not
always straightforward. Only a few reactions have been reported in the literature where
a modification is applied directly on the β-NAD+ scaffold,
70, 83, 107-109 generally
involving the introduction of substituents on the nucleobases or the riboses. However,
because of the extreme fragility of β-NAD+, the most common method to obtain NAD
+
derivatives substituted on the nucleobases, the riboses or the pyrophosphate bond,
consists of the introduction of a modification on a stable precursor of the β-NAD+
scaffold, such as the nucleoside/nucleotide moieties, and the following synthesis of the
NAD+ target.
As discussed in Chapter 1 (see 1.5), the project aim was the synthesis of NAD+
derivatives modified on C-2, C-6 or C-8 of the adenine ring with aryl/heteroaryl groups,
and their following biological evaluation toward NAD+-consuming enzymes. Therefore,
the synthesis of adenine-modified NAD+ derivatives was explored. The most common
pathway toward adenine-modified NAD+ derivatives is the multistep synthesis (pathway
a, Scheme 2.1), which starts with the substitution of the adenosine in the desired
Chapter 2
51
position, subsequently undergoing the phosphorylation of 5′-OH, the phosphate
activation and the final coupling with β-NMN to generate the pyrophosphate bond.110-112
Alternative routes were explored to shorten the synthesis of adenine-modified NAD+
derivatives: the pathway b (Scheme 2.1) via nucleoside phosphoramidite intermediates
gives in one step the phosphorylation/phosphate activation of substituted adenosines for
their following coupling with β-NMN;113
the pathway c (Scheme 2.1) allows
bromination and substitution by Suzuki-Miyaura cross-coupling in one-pot, on position
8 of the natural β-NAD+.70
Scheme 2.1 General synthetic pathways to aryl/heteroaryl adenine-modified NAD+ derivatives. Pathway
a) via AMP-morpholidates; pathway b) via nucleoside phosphoramidites; pathway c) via one-pot, two-
step synthesis of 8-substituted NAD+ derivatives. Abbreviations used: β-NAD
+, nicotinamide adenine
dinucleotide; AMP, adenosine monophosphate; β-NMN, nicotinamide mononucleotide.
The pathway a, though more laborious and time-consuming, is the most flexible to
successfully obtain different substitutions on the β-NAD+ scaffold, since even the
modifications requiring harsh reaction conditions are introduced on the stable adenosine
precursors. Indeed, this pathway was used to synthesize all the final NAD+ derivatives
Chapter 2
52
studied in this thesis. Through substitution of the adenosine on C-2, C-6 or C-8 of the
adenine ring and phosphorylation of the 5′-OH, the NAD+ multistep synthesis converges
to two final steps, in common for all the substituted adenosines, leading to the final
NAD+ derivatives. This route will be discussed first, starting with the synthesis of the
differently adenine-modified AMP derivatives (see 2.3, 2.4 and 2.5), converging then to
the common steps to obtain the final NAD+ derivatives (see 2.6). The pathway b and c
will be explored as well, though subject to many future further studies (see 2.7).
2.2 Substitution of the adenine ring by palladium-mediated cross-couplings
Since the 1940s and 1950s, the study of nucleosides, nucleotides and their derivatives
has become more and more important, in parallel with the improved and deeper
knowledge of DNA and metabolic processes involving these compounds. Their several
biological roles have made them a constant source of interest as therapeutics and
molecular probes to understand biological systems. From the first attempts to modify
the natural substrates on the nucleobase or the ribose, to the most successful attempts to
substantially change the structure, improving the derivatives stability and selectivity, an
extensive study to new approaches for these modifications has been undertaken
resulting in the development of many useful synthetic pathways.114
Since this project aimed to synthesize of NAD+ derivatives modified on the adenine
ring, attention was mainly focused on the modified purines, also found in nature, well
known for their extremely various and important biological properties.115
Several have
therapeutic activity as modulators of adenosine receptors or stimulators of plant cell
growth and division (cytokinines); others inhibit DNA polymerases exhibiting antiviral,
antibacterial, anticancer properties. Moreover, purines modified by metabolically
activated mutagens and carcinogens introduced in the cells by dietary and
environmental exposure, can cause DNA alteration, and even oncogenesis. Indeed, the
modification of the purine ring in different positions can substantially alter the normal
nucleoside/nucleotide conformation, the H-bonding ability and, consequently, the
intercalation in DNA and the interaction with proteins.
Chapter 2
53
Figure 2.1 Purine structure and numbering.
The purine ring, which skeleton is given by the fusion of imidazole and pyrimidine
rings (Figure 2.1), can be substituted on C-2, C-6 and C-8 and on N-1, N-3, N-7 and N-
9.116
Substituted purines can be obtained by different methods:
Heterocyclization from pyrimidine or imidazole; however, in particular cases the
methods are not efficient and require many reaction steps with quite low yield;115, 117
SNAr, where halogenated electrophilic purines can undergo displacement of the
halogen with a good nucleophile;118
SEAr, where nucleophilic purines can be substituted with electrophilic reagents (e.g.
alkylation).119
The last two methods require particular electronic properties and drastic reaction
conditions; consequently, other easier synthetic pathways have been explored. The
recent development of metal-mediated couplings has considerably facilitated the
synthesis of purine derivatives.114, 120, 121
In general, metal-catalyzed couplings are
economically advantageous; in addition, they are suitable for introducing a broad range
of substituents. A wide variety of palladium-mediated reactions, such as Sonogashira,
Heck, Stille and Suzuki-Miyaura cross-couplings, has been developed since the 1960s,
and adapted and optimized to prepare respectively alkynyl, alkenyl and aryl/heteroaryl
purine derivatives.114, 116, 120-127
The synthetic part of this project was focused on the substitution of C-2, C-6 or C-8 of
the adenine ring in adenosine and/or adenosine monophosphate with aryl/heteroaryl
groups, as intermediates to synthesize the final targets, the corresponding adenine-
modified NAD+ derivatives. In particular, the introduction of phenyl and pyrrolyl
groups in different positions on the adenine ring was considered in order to screen and
Chapter 2
54
compare eventual structural and biological differences caused by the hydrophobic
interactions exerted by the phenyl group, with those due to H-bonding formation of the
pyrrolyl group. In addition, the substitution of the adenine with a pyrrolyl group could
be particularly interesting because of the presence of pyrrole units in many biological
molecules.
Suzuki-Miyaura and Stille cross-couplings were the most suitable palladium-mediated
reactions for our purposes. They catalyze respectively the C-C bond formation between
aryl/heteroaryl boronic acids, XB(OH)2, or stannanes, XSnBu3, and activated aromatic
halides (Scheme 2.2). Owing to the commercial availability and relatively easy
halogenation of purine nucleosides/nucleotides (1-3), the general cross-coupling
strategy uses the halogenated nucleosides/nucleotides in the favoured position as
starting materials (4-6), followed by the coupling with boronic acids (a) or stannanes (b)
via palladium-mediated reactions. Palladium-mediated cross-couplings can be
performed directly on adenosine, avoiding the halogenation step, although this generally
requires harsh reaction conditions; only recently, mild reaction conditions have been
found, also suitable for the direct substitution of the unstable and unprotected 2′-
deoxyadenosine.128-130
Scheme 2.2 General synthetic strategy to aryl/heteroaryl adenine-modified nucleotides. Guanosine (1) as
starting material for 2-substituted adenosine monophosphate (7); inosine (2) as starting material for 6-
substituted purine nucleotide (8); adenosine (3) as starting material for 8-substituted adenosine
monophosphate (9). After the halogenation of the nucleosides, 1-3, in the favoured positions, 4-6 undergo
Suzuki-Miyaura (a) or Stille cross-coupling (b) by Pd/lig catalyst (lig = phosphine) to introduce the
aryl/heteroaryl group, X. The following phosphorylation generates the corresponding nucleotides, 7-9.
Chapter 2
55
The Suzuki-Miyaura131
and Stille132
cross-couplings proceed via a catalytic cycle,
involving an oxidative addition, transmetallation and reductive elimination (Scheme
2.3). In the first step, the reactive species, Pd(0), is oxidized to Pd(II) after the addition
of an organic halide, R-X, forming a cis complex, that rapidly isomerizes to trans: this
is generally the rate-limiting step in the catalytic cycle, depending on the reactivity of
the halide and the ζ-donating ability of the ligands coordinating Pd(0). During the
transmetallation step, the second organic group, R′, is transferred to Pd(II) from another
metallic species. In the Suzuki-Miyaura cross-coupling this is a boronic acid, while in
the Stille cross-coupling it is an organic stannane. The use of a base in the Suzuki-
Miyaura cross-coupling is fundamental to activate the organic boronic acid during the
transmetallation step; indeed, the organic group on the boron, because of the
predominantly covalent B-C bond, would not otherwise be nucleophilic enough to take
part in the transmetallation step. In contrast, the organic stannanes do not need any
activation because of the weaker Sn-C bond. The last step in the catalytic cycle, the
reductive elimination, requires a slow isomerization of the trans Pd(II) complex to cis
with elimination of the cross-coupled product, R-R′, and the re-establishment of the
reactive species, Pd(0).
Scheme 2.3 Catalytic cycle of palladium-mediated cross-couplings.
Chapter 2
56
The toxicity of stannanes and their difficult disposal, compared to the high stability,
easy handling, low toxicity and huge variety of available boronic acids has favoured the
development of the Suzuki-Miyaura cross-coupling compared to the Stille cross-
coupling. Nevertheless, in some cases only the Stille cross-couplings are really effective
for the formation of certain types of C-C bonds.131-133
2.3 Synthetic route to 2-aryl/heteroaryl AMP derivatives
The synthetic route to 2-aryl/heteroaryl AMP derivatives starts with the halogenation of
C-2 on the adenine ring in order to form the organic halide, which will undergo the
Suzuki-Miyaura or Stille cross-coupling for the introduction of the aryl/heteroaryl
substituent.
The most common substitutions on C-2 are normally achieved by halogenation of this
position, followed by nucleophilic aromatic substitutions, formation of palladium-
catalyzed C-N bonds, or Sonogashira cross-couplings; the latter is an alternate copper-
catalyzed palladium-mediated cross-coupling to obtain 2-alkynyl adenine derivatives.134,
135 Very few examples of Suzuki-Miyaura and Stille cross-couplings on C-2 of
adenosine are present in the literature.127, 136-138
Most of these substitutions start from
guanosine (1) and proceed via the installation of an iodine on C-2, as this generates the
most reactive halide (4) (Scheme 2.2). Indeed, the substitutions on C-2 of the adenine
ring are complicated by the low reactivity of this position, which is on the electron-poor
pyrimidine ring of the purine skeleton; hence, the introduction of a very good leaving
group is necessary. To date, the cross-coupling reactions described in the literature for
the 2-substitution have only been applied on the adenosine, but not on the adenosine
monophosphate. Indeed, these reactions are carried out in organic phase with apolar
reagents and solvents, requiring for this reason the protection of the hydroxyl groups of
adenosine with different protecting groups.127, 136-138
Chapter 2
57
Our synthetic strategy aimed at the synthesis of 2-iodo AMP (7a), 2-phenyl AMP (7b)
and 2-(N-Me-pyrrol-2-yl) AMP (7c). It proceeded through the iodination of adenosine
on C-2 (4) and the protection of the 2′, 3′-OH on the ribose with acetone to form the
isopropylidene protecting group (4a). After the Pd-mediated cross-couplings for the
introduction of phenyl and N-Me-2-pyrrolyl groups (4a-c), the phosphorylation of 5′-
OH via phosphoramidite intermediates (4d-f) or by POCl3 afforded 4g-i, finally
deprotected to give the final compounds (7a-c) (Scheme 2.4).
Scheme 2.4 Synthetic route to 2-substituted AMP derivatives. (i)-(v) Iodination of adenosine on C-2 from
guanosine (1); (vi) 2′,3′-O-isopropylidene protection; (vii) palladium-mediated cross-coupling; (viii)
phosphorylation of 5′-OH; (ix) deprotection of 2′,3′-OH.
2.3.1 Iodination of C-2 and 2′,3′-O-isopropylidene protection of adenosine
The first step in the synthesis of 2-substituted AMP derivatives is the iodination of C-2
in adenosine, particularly required to have the most reactive halide in the following
Suzuki-Miyaura and Stille cross-couplings. The halogenation of aromatic rings is
generally achieved by SEAr; in the case of the adenine ring, the halogenation in position
2 is less straightforward than that in position 8 because of a lower reactivity of C-2
Chapter 2
58
compared to C-8 in SEAr. For this reason, the iodination of C-2 is generally achieved by
a long synthetic strategy, well described in the literature, which avoids the use of SEAr.
It starts from guanosine (1), converting it to the fully acetylated 2-amino-6-chloro
purine nucleoside (1a); the latter is converted by diazotization-iodine substitution in the
fully acetylated 2-iodo-6-chloro purine nucleoside (1b), which undergoes amination and
deprotection to finally give the 2-iodo adenosine (4) (Scheme 2.5).134, 139, 140
Scheme 2.5 Iodination of adenosine on C-2 from 1. (i) Acetic anhydride, DMF, pyridine, 75 °C, 4 h,
87%; (ii) POCl3, N,N-dimethylaniline, 100 °C, 10 min, 85%; (iii) acetic anhydride, DMAP, TEA,
acetonitrile, rt, 1 h, 89%; (iv) CuI, I2, CH2I2, isoamyl nitrite in THF, 70 °C, 4 ‒ 12 h, 85% from 1a, 51%
from 1c; (v) NH3/MeOH, 60 °C, 12 h, 40% from 1b, 30% from 1d; 33% overall from 1a, 21% overall
from 1c.
Our synthesis, described in Scheme 2.5, started from the commercially available 2-
amino-6-chloro purine nucleoside (1c), which was fully protected by acetylation to
obtain 1a (89%). Afterwards, the next reactions were carried out on the protected
nucleosides as described in the literature, until the final amination/deprotection step,
which gave compound 4 with 33% yield overall (compared to 70% yield in the
literature).139
The same pathway was performed on the unprotected nucleosides, starting
with the diazotization-iodine substitution of 1c to give 1d (51%), finally converted in
the compound 4 with 21% yield overall. Therefore, this pathway, although giving
product 4 in lower yields, allowed to avoid the protection/deprotection steps.
Chapter 2
59
Afterwards, the obtained 2-iodo adenosine (4) had to be protected on the ribose to
undergo the subsequent palladium-mediated cross-coupling in organic phase. Because
of our synthetic strategy, the isopropylidene protecting group was the most convenient
to install. Indeed, because of its acetal nature, it was resistant to the basic conditions of
the Suzuki-Miyaura cross-coupling, and allowed the protection of the vicinal hydroxyl
groups in C-2′ and C-3′, leaving the primary alcohol in C-5′ free for the subsequent
phosphorylation.
The 2′,3′-O-isopropylidene protecting group can be introduced by reaction of the
nucleoside with acetone in acidic conditions (Scheme 2.6);135, 141
the acetal formation is
a reversible equilibrium, hence the deprotection occurs again in acidic conditions. The
2′,3′-O-isopropylidene protection of 4 was carried out as described in the literature,
affording the desired product 4a with 98% yield.
Scheme 2.6 2′,3′-O-isopropylidene protection of 2-iodo adenosine, 4. (vi) Acetone, 70% HClO4, rt, 1 h;
then, NH4OH, rt, 3 h, 98%.
2.3.2 Suzuki-Miyaura and Stille cross-couplings on C-2 of 2′,3′-O-isopropylidene-2-iodo adenosine
As mentioned in 2.3, the reactivity of C-2 on the adenine ring toward palladium-
mediated cross-couplings is lower compared to other positions, due to the electron-poor
nature of the pyrimidine ring; indeed, the electrophilicity of the organic iodide is
decreased because of a strong mesomeric effect competing with the inductive effect.
Chapter 2
60
In the literature, the optimized conditions for the Suzuki-Miyaura cross-coupling on the
protected 2-iodo adenosine (4a) allow the successful coupling with simple aryl boronic
acids,127, 136-138
such as phenyl or biphenyl boronic acids; Pd(OAc)2, (2-
biphenyl)dicyclohexylphosphine, K3PO4, in anhydrous 1,4-dioxane at 100 °C is the
optimal catalytic system, affording the desired products in 60% yield.136
In this project,
these conditions using phenyl boronic acid were successfully applied to obtain the 2′,3′-
O-isopropylidene-2-phenyl adenosine (4b) in 43% yield (Scheme 2.7).
No references in the literature have been found on Suzuki-Miyaura cross-couplings at
C-2 with heteroaryl boronic acids. The installation of a pyrrolyl group on the adenine
ring by Suzuki-Miyaura cross-coupling would require the use of 2-pyrrole boronic acid.
However, the latter is a difficult substrate because it can dimerize or protodeboronate,
and the free amino group can coordinate Pd(0) inhibiting the catalytic cycle142, 143
. The
protection of the amino group of the pyrrole with different protecting groups (Boc,
TIPS) has solved these problems in many cases affording the pyrrolyl cross-coupled
products with quite good yields. Nevertheless, the Suzuki-Miyaura cross-coupling on C-
2 of the 2′,3′-O-isopropylidene-2-iodo adenosine (4a) with the N-Boc-2-pyrrole boronic
acid failed using the optimized conditions for the phenyl substitution.136
The only example found in the literature for the synthesis of 2-heteroaryl substituted
protected adenosine has used the Stille cross-coupling to introduce thienyl and furanyl
groups137
on the fully tert-butyldimethylsilyl protected 2-iodo adenosine. In the context
of this project, the commercially available tributyl-(N-Me-pyrrol-2-yl) stannane was
used to test the Stille cross-coupling conditions on 2′,3′-O-isopropylidene-2-iodo
adenosine (4a), requiring PdCl2(PPh3)2 as catalyst. Effectively, under these conditions,
using anhydrous 1,4-dioxane instead of anhydrous THF, about 43% of 4a was
converted in 4c, slightly improved to 50% when anhydrous LiCl was added to the
reaction; indeed, the addition of chloride in Stille cross-couplings can often accelerate
the catalytic reaction during the transmetallation step (Scheme 2.7). However, the
purification by silica gel chromatography of the product 4c from the excess of stannane
was unsuccessful, causing a substantial stannane contamination of the reaction product
even after several purifications; therefore, the next synthetic steps were performed on
the crude mixture.
Chapter 2
61
Scheme 2.7 Palladium-mediated cross-couplings on C-2 of 2′,3′-O-isopropylidene-2-iodo adenosine, 4a.
Suzuki-Miyaura cross-coupling (1): phenyl boronic acid, Pd(OAc)2, (2-biphenyl)dicyclohexylphosphine,
K3PO4 in anhydrous 1,4-dioxane, 100 °C, 24 h, 43% for 4b. Stille cross-coupling (2): tributyl-(N-Me-
pyrrol-2-yl) stannane, optional anhydrous LiCl, PdCl2(PPh3)2 in anhydrous 1,4-dioxane, 100 °C, 48 h, 43-
50% conversion for 4c.
2.3.3 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine and subsequent deprotection to 2-aryl/heteroaryl AMP derivatives
A common method to phosphorylate the 5′-OH of protected and/or unprotected
nucleosides uses POCl3 as phosphorylating agent (method 1). In the past, many studies
have been performed to optimize this reaction affording quite good yields, even in the
difficult case of the phosphorylation of unprotected nucleosides, where the selective
phosphorylation of 5′-OH is required. The optimized reaction conditions require the
addition of a base to the reaction mixture to buffer the acidic pH generated along the
reaction course; indeed, adenosine is subjected to depurination under acidic conditions.
Moreover, the solvents commonly used are acetonitrile or triethylphosphate, the latter
accelerating the phosphorylation in a selective way towards the 5′-OH.144-146
In this study, when the phosphorylation was performed on 2′,3′-O-isopropylidene-2-
phenyl adenosine (4b) using POCl3, proton sponge as a base and triethylphosphate as
solvent, the conversion was so low that, in some cases, no product could be recovered.
In addition, when the reaction failed, the recovery of the starting material was
complicated because of the difficult purification from triethylphosphate and proton
sponge. The phosphorylation by POCl3 was also attempted on 2′,3′-O-isopropylidene-2-
iodo adenosine (4a), substituting the triethylphosphate with acetonitrile and removing
the proton sponge to facilitate the purification step. However, a poor conversion was
observed also in this case, probably due to a substantial substitution of the iodide with
the chloride generated in the course of the reaction (Scheme 2.8).
Chapter 2
62
Because of the unsatisfactory results obtained in the phosphorylation of 4a and 4b by
POCl3, this phosphorylation method was not applied on the crude 2′,3′-O-
isopropylidene-2-(N-Me-pyrrol-2-yl) adenosine (4c).
Scheme 2.8 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine by POCl3
(method 1). (viii) Optional proton sponge, triethylphosphate or anhydrous acetonitrile, 0 °C, 12 h, 9%
from 4a; (ix) Dowex 88 (H+), rt, 12 h, 36% for 7a; 3% overall for 7a.
Therefore, an alternative phosphorylation method was chosen using the
phosphoramidite as a highly reactive phosphorylating agent (method 2).113
The new
phosphorylation strategy required the reaction of 2′,3′-O-isopropylidene-2-substituted
adenosines (4a-c) with 2-cyanoethyl N,N-diisopropylchlorophosphoramidite to obtain
the phosphoramidite nucleoside intermediates (4d-f). These could then be oxidized to
the phosphate intermediates by tBuOOH, and deprotected from 2-cyanoethyl and N,N-
diisopropyl groups by treatment with NH4OH, to afford the 2′,3′-O-isopropylidene-2-
substituted adenosines 5′-monophosphate (4g-i). After removal of the O-isopropylidene
protecting group by treatment with Dowex 88 (H+), the final 2-substituted AMP
derivatives (7a-c) were obtained (Scheme 2.9).
This synthetic route, though involving more steps compared to the POCl3 method and
the use of an expensive reagent in the form of 2-cyanoethyl N,N-
diisopropylchlorophosphoramidite, turned out to be successful in the phosphorylation of
2′,3′-O-isopropylidene-2-substituted adenosines (4a-c), affording 7a and 7b in 19% and
17% yield overall respectively.
Chapter 2
63
Unfortunately, 7c was never recovered pure, even after several purifications, due to the
stannane contamination from the previous Stille cross-coupling.
Scheme 2.9 Phosphorylation of 5′-OH of 2′,3′-O-isopropylidene-2-substituted adenosine via
phosphoramidite (method 2). (viii) DIPEA, anhydrous CH2Cl2, rt, 1 h; then, 5.0 – 6.0 M tBuOOH in
nonane, rt, 1 h; NH4OH, rt, 3 h; (ix) Dowex 88 (H+), rt, 12 h; 19% overall for 7a, 17% overall for 7b;
yield not estimated for 7c.
As will be discussed in 2.7.1, the phosphoramidite strategy could be used to form the
pyrophosphate bond between the activated phosphite intermediates and β-NMN, to
generate the final NAD+ targets. However, this route still needs further investigations
and optimizations.
2.4 Synthetic route to 6-aryl/heteroaryl AMP derivatives
The synthetic strategy to 6-phenyl and 6-(pyrrol-2-yl) AMP derivatives, respectively 8a
and 8b, started with the Suzuki-Miyaura cross-coupling on the commercially available
6-chloro purine nucleoside (5), that could be otherwise readily synthesized from inosine
(2) by treatment with POCl3 and N,N-diethylaniline.140
To avoid the difficult
phosphorylation step on unprotected nucleosides, the 6-substituted purine nucleosides
(5a-b) were protected on the 2′, 3′-OH by O-isopropylidene protecting group (5c-d).
Then, these intermediates were phosphorylated on the 5′-OH via phosphoramidite
intermediates (5e-f) or by POCl3 to afford 5g-h, then finally deprotected to obtain the 6-
substituted AMP derivatives (8a-b) (Scheme 2.10).
Chapter 2
64
Scheme 2.10 Synthetic route to 6-aryl/heteroaryl AMP derivatives. (i) Suzuki-Miyaura cross-coupling;
(ii) 2′,3′-O-isopropylidene protection; (iii) phosphorylation of 5′-OH; (iv) deprotection of 2′,3′-OH.
2.4.1 Suzuki-Miyaura cross-coupling on C-6 of 6-chloro purine nucleoside
The Suzuki-Miyaura cross-coupling on C-6 of purine nucleosides is well described in
the literature and remarkable improvements have been achieved in the optimization of
its conditions. The first examples of substitutions on C-6 concerned the cross-coupling
of simple substituted aryl groups on the protected purine nucleosides.147
Since the
unprotected compounds showed cytostatic activities, the scientific interest toward these
substitutions enhanced. Later, the attention was focused on the introduction of
heteroaryl groups that exhibited, in addition to the cytostatic activities, strong anti-HCV
(Hepatitis C Virus) activities.148
Hence, a large number of heteroaryl substituents have
been introduced on the fully protected purine nucleosides using different methods, such
as the Suzuki-Miyaura cross-coupling, the heterocyclization by 1,3-dipolar
cycloadditions, or the palladium-mediated Negishi cross-coupling, involving organozinc
reagents.148, 149
More recently, Suzuki-Miyaura cross-couplings on the unprotected 6-
chloro purine nucleoside (5) have been successfully carried out in the aqueous phase for
Chapter 2
65
the introduction of some heteroaryl or unprotected phenylalanine substituents.149, 150
However, no references have been found for the Suzuki-Miyaura cross-coupling on the
unprotected 6-chloro purine nucleoside (5) in an aqueous phase with N-Boc-2-pyrrole
boronic acid.
In this project, the aqueous cross-coupling conditions described in the literature for the
synthesis of unprotected 8-substituted adenosine and guanosine, requiring Na2PdCl4,
TPPTS and K2CO3 as catalytic system,144, 151, 152
were applied to the synthesis of the 6-
phenyl (5a) and 6-(pyrrol-2-yl) (5b) purine nucleosides, affording the desired products
in 50% and 37% yield respectively (Scheme 2.11). Interestingly, during the synthesis of
5b, the Boc-deprotection of the pyrrolyl substituent occurred directly under the basic
cross-coupling conditions.153, 154
Scheme 2.11 Suzuki-Miyaura cross-coupling on C-6 of 6-chloro purine nucleoside, 5. Phenyl or N-Boc-2-
pyrrole boronic acid, Na2PdCl4, TPPTS, K2CO3 in water, 100 °C, 1 h (5a), 24 h (5b); 50% for 5a, 37%
for 5b.
Substantial differences in polarity between the phenyl (5a) and the pyrrolyl (5b)
substituted purine nucleosides were noticed during their purifications. Indeed, 5a was
isolated and purified from the reaction mixture by extraction in organic phase and
following recrystallization from water. By contrast, 5b was soluble in water and needed
to be purified by ion-pair chromatography. Moreover, the reaction mixture from the
cross-coupling with the N-Boc-2-pyrrole boronic acid was quite complex, and only a
very slow gradient during the ion-pair purification allowed the separation between a
multitude of different peaks. These were not isolated and analysed, but some of them
could potentially be attributed to the dimer of the pyrrole,142
or to the N-arylated
Chapter 2
66
product, formed by coupling of the nitrogen of the pyrrole with the chlorine of the 6-
chloro purine nucleoside.149
2.4.2 2′,3′-O-isopropylidene protection and phosphorylation of 5′-OH of 6-substituted purine nucleosides, with subsequent deprotection to 6-aryl/heteroaryl AMP derivatives
The 2′,3′-O-isopropylidene protection of 6-substituted purine nucleosides (5a-b) was
carried out with the same conditions described in 2.3.1, affording 5c-d with 85% and
39% yield respectively. Afterwards, the phosphorylation of 5′-OH of 2′,3′-O-
isopropylidene-6-substituted purine nucleosides (5c-d) was performed by POCl3
(method 1) and via the phosphoramidite (method 2), as previously described in 2.3.3
(Scheme 2.12).
Scheme 2.12 Phosphorylation of 5′-OH and following deprotection of 2′,3′-O-isopropylidene-6-
substituted purine nucleosides. (iii) Method 1 by POCl3: POCl3, anhydrous acetonitrile, 0 °C, 12 h. (iii)
Method 2 via phosphoramidite: 2-cyanoethyl N,N-diisopropylchlorophosphoramidite, DIPEA, anhydrous
CH2Cl2, rt, 1 h; then, 5.0 – 6.0 M tBuOOH in nonane, rt, 1 h; NH4OH, rt, 3 h. (iv) (Method 1): Dowex 88
(H+), rt, 12 h, 30% for 8a. (iv) (Method 2): AcOH/H2O (1:9), 90 °C, 3 h, complete conversion for 8b;
28% for 8a and 50% for 8b overall via phosphorylation method 1; 7% for 8a overall via phosphorylation
method 2.
Chapter 2
67
The phosphoramidite method, selected for the phosphorylation of 2′,3′-O-
isopropylidene-2-substituted adenosines, allowed the phosphorylation of the 2′,3′-O-
isopropylidene-6-phenyl purine nucleoside (5c), affording the desired product 8a with
7% yield overall. By contrast, when the same phosphorylation method was applied to
the 2′,3′-O-isopropylidene-6-(pyrrol-2-yl) purine nucleoside (5d), no product was
recovered after the different deprotection steps. In particular, the final treatment with
Dowex 88 (H+) to remove the O-isopropylidene protecting group on the ribose seemed
to be responsible for the loss of the material. Indeed, 5h, especially if present in small
amounts, could have remained attached to resin; hence, the use of a weak acid cation
exchange resin could have allowed an easier and more efficient recovery of the material.
Owing to the poor yields obtained for the phosphorylation via phosphoramidite (method
2), the method with POCl3 (method 1) was reconsidered and slightly modified to allow,
in case of failure, the recovery of the starting materials. Therefore, the reactions were
performed in anhydrous acetonitrile without the additional proton sponge to simplify the
purification step. Under these conditions, a partial deprotection of the O-isopropylidene
protecting group occurred because of the acidic pH developed during the reaction
course. Then, the complete O-isopropylidene deprotection of 5g was carried out by the
usual treatment with Dowex 88 (H+). However, a new deprotection method was used for
5h, requiring its treatment with an acetic acid solution at 90 °C for 3 h, yielding a
complete deprotection. Compounds 8a-b were obtained in 28% and 50% yield overall
respectively (Scheme 2.12).
2.5 Synthetic route to 8-aryl/heteroaryl AMP derivatives
The synthetic strategy to 8-aryl/heteroaryl AMP derivatives started with the
bromination of adenosine (3) or adenosine monophosphate (3a) in position 8 of the
adenine ring; this reaction is selective for C-8, as this position is very reactive to SEAr,
as mentioned in 2.3.1.155, 156
The subsequent Suzuki-Miyaura cross-coupling in position 8 is very well documented
in the literature and a large variety of aryl/heteroaryl groups have been introduced in
this position. The development of hydrophilic ligands and palladium sources has
Chapter 2
68
allowed its performance in aqueous phase with unprotected nucleosides/nucleotides,
avoiding time-consuming protection and deprotection steps.144, 150-152, 157-159
Two routes can be followed to obtain the final 8-aryl/heteroaryl AMP derivatives (9)
(Scheme 2.13). The longest route starts from adenosine (3), via its bromination in
position 8 to give 6. The latter can alternatively undergo the phosphorylation of 5′-OH
by POCl3 to obtain 9a, or the Suzuki-Miyaura cross-couplings to obtain the 8-
aryl/heteroaryl adenosine derivatives. These intermediates can be further
phosphorylated on 5′-OH to give the final 8-aryl/heteroaryl AMP derivatives (9). The
shortest route starts from adenosine monophosphate (3a), via its bromination in position
8 to give 9a. The latter can directly undergo the Suzuki-Miyaura cross-coupling to give
the final 8-aryl/heteroaryl AMP derivatives (9). In this project, the choice to follow the
shortest synthetic route led in a short period of time to the synthesis of a small library of
8-aryl/heteroaryl AMP derivatives (9b-n).
Scheme 2.13 Synthetic route to 8-aryl/heteroaryl AMP derivatives. (i) Bromination on C-8; (ii) Suzuki-
Miyaura cross-coupling; (iii) phosphorylation of 5′-OH (R is an aryl/heteroaryl group).
Chapter 2
69
2.5.1 Bromination of C-8 of adenosine and adenosine monophosphate
The bromination on C-8 of adenosine (3) has been performed by adding saturated
bromine water to the adenosine solution in sodium acetate buffer (pH 4). The buffered
conditions are fundamental to avoid the pH dropping down dramatically because of the
release of HBr from the reaction, with consequent depurination. Under these conditions,
the product 6 crashed out from the reaction mixture as a solid in 80% yield.155, 156
The bromination on C-8 of adenosine monophosphate (3a) described in the literature
requires the same conditions as for the bromination of adenosine,155, 156
starting from its
sodium salt. Slight modifications were performed in this study to simplify this
procedure, using neat bromine instead of saturated bromine water. Since the formation
of side-products was observed during the reaction in sodium acetate buffer, the solvent
was substituted with a 0.25 M KH2PO4 solution (pH 4). An almost complete conversion
of 3a to 9a was obtained with 50-60% yield after ion-pair chromatography purification
(Scheme 2.14).
Scheme 2.14 Bromination on C-8 of adenosine, 3, and AMP, 3a. Method 1: 3, saturated bromine water, 1
M NaOAc buffer (pH 4), rt, 24 h, 80%. Method 2: 3a, neat bromine, 0.25 M KH2PO4 (pH 4), rt, 24 h, 50-
60%.
Chapter 2
70
2.5.2 Suzuki-Miyaura cross-coupling on C-8 of 8-bromo adenosine monophosphate to 8-aryl/heteroaryl AMP derivatives
The Suzuki-Miyaura cross-coupling on C-8 of purine nucleosides/nucleotides has been
investigated in detail and successful conditions are well established in the literature,
using Na2PdCl4, TPPTS and K2CO3 in water as the optimal catalytic system.144, 150-152,
157-160
This project was aimed in particular at the challenging synthesis of N-heteroaryl or
amino-aryl substituted AMP derivatives. Indeed, the cross-coupling reactions with free
amino moieties are generally complicated by the inhibition of the catalytic center by N-
coordination, or polymerization, in addition to the low reactivity of electron-deficient N-
heteroaryl boronic acids.142, 143, 161, 162
Therefore, an accurate optimization of the
reaction conditions was required to successfully afford the desired products.
A small library of 8-aryl/heteroaryl AMP derivatives (9b-n) (Table 2.1, Scheme 2.15)
was generated and their conformational and spectroscopic properties investigated in
Chapter 3. A range of aryl, electron-poor and electron-rich substituents was introduced
in position 8 of the adenine ring. The Suzuki-Miyaura cross-coupling conditions were
optimized to obtain decent yields even for the most difficult substitutions with electron-
deficient N-heteroaryl groups.
cmpd R (substituent on C-8) Yield (%)a
9b Phenyl 80-90
9c 3-(N-Boc-aminomethyl)phenyl 78
9d 3-(Aminomethyl)phenyl 85
9e Pyridin-3-yl 70
9f Pyridin-4-yl 80
9g 2,4-DMT-pyrimidin-5-yl 75
9h Pyrrol-2-yl 57
9i Furan-2-yl 67
9j 1H-Pyrazol-4-yl 57
9k 5-Formyl-thien-2-yl 52
9l 5-Bromo-thien-2-yl 17
9m Indol-2-yl 83
9n Benzofuran-2-yl 10 aYield of isolated products
Table 2.1 Synthesis of the 8-aryl/heteroaryl AMP derivatives (9b-n).
Chapter 2
71
Scheme 2.15 Suzuki-Miyaura cross-coupling on C-8 of 8-bromo AMP, 9a. (ii) Aryl/heteroaryl boronic
acid, Na2PdCl4, TPPTS, K2CO3 in water, Δ, 1 – 24 h, 10-90%.
Generally, the Suzuki-Miyaura cross-couplings with simple aryl boronic acids can be
performed at low temperature using small amounts of base and palladium/ligand
catalyst. For example, 9b and 9c could be synthesized at only 40 °C using the right
catalytic system. In contrast, the Suzuki-Miyaura cross-couplings with electron-poor
boronic acids generally require stronger reaction conditions, using very high
temperature and more equivalents of base and palladium/ligand catalyst to react in the
transmetallation step.
In this study, the pyridinyl group was the most electron-poor in the range of selected
substituents to introduce on the position 8 of the adenine ring. Due to the electron-poor
nature of the pyridine ring, the boron in the 3-pyridinyl boronic acid has a pronounced
Lewis acid behavior, which makes it poorly reactive. However, in water the 3-pyridinyl
boronic acid is normally present as a zwitterion, with a positive charge on the pyridine
basic nitrogen and a negative charge on the boron (Figure 2.2). Therefore, the aqueous
conditions increase the reactivity of the boron and block the pyridine basic nitrogen
from inhibiting the catalytic palladium centre.162, 163
Chapter 2
72
Figure 2.2 Ionization equilibrium of the 3-pyridinyl boronic acid in water.
Due to its poor reactivity, the Suzuki-Miyaura cross-coupling of 9a with the 3-pyridinyl
boronic acid to obtain 9e, required high temperature (100 °C) and a Pd/ligand ratio 1:3.
A screening of base equivalents, ranging from 1.5 to 10, showed a slow conversion with
1.5 equiv., fast with 3 and 5, but no conversion with 10 equiv. of base. The application
of micellar-catalyzed Suzuki-Miyaura cross-coupling was considered to perform the
reaction at lower temperature in water and in absence of co-solvent. Indeed, the
formation of micelles should improve the solubility of the boronic acid in water
accelerating the reaction.164-166
In particular, the use of surfactants, such as PTS
(polyoxyethanyl α-tocopheryl sebacate, a nonionic amphiphile) and TBAF, was
investigated, performing parallel reactions between 9a and the 3-pyridinyl boronic acid
at 40 °C for 2 days to obtain 9e. While the reactions did not work with the standard
amounts of Pd catalyst (2 – 2.5 mol%), adding more (10 mol%), a conversion was
observed, higher in the reactions performed with PTS (> 60%) than those with TBAF (≤
40%).
Several interesting observations were made during the synthesis of 9c and 9d. When the
Suzuki-Miyaura cross-coupling was performed with the 3-(N-Boc-aminomethyl)phenyl
boronic acid at 80 °C in 1 h, no Boc-deprotection occurred in the basic reaction
conditions. By contrast, when the same reaction was performed at 40 °C over two days,
an almost complete deprotection occurred, affording 9d. To obtain the unprotected 9d,
two different synthetic routes were used (Scheme 2.16). In the first istance, the
deprotection of the cross-coupled 9c was carried out by refluxing with TFA in methanol
until the disappearance of the starting material was observed. In the second route, the
deprotection of the 3-(N-Boc-aminomethyl)phenyl boronic acid with TFA in CH2Cl2 at
room temperature was performed to use the unprotected boronic acid in the subsequent
Suzuki-Miyaura cross-coupling with 9a to afford 9d.
Chapter 2
73
Scheme 2.16 Synthesis of 9d. (i) TFA, MeOH, reflux; (ii) TFA, CH2Cl2, rt, 1 h, quantitative yield; (iii)
Suzuki-Miyaura cross-coupling: 9a, Na2PdCl4, TPPTS, K2CO3 in water, 80 °C, 2 h.
With the first strategy, after the disappearance of the starting material 9c, two new spots
could be observed by TLC (thin layer chromatography), one very polar and one less
polar than the starting 9c. Probably, a mixture of the differently protonated forms of 9d,
in equilibrium with each other during the TLC run in iPA/H2O/NH4OH (6:3:1), was
observed. After purification by ion-pair chromatography, 9d was recovered as a single
product corresponding to the most polar spot in TLC.
A titration of the obtained 9d was performed with NaOH and HCl to investigate its
different protonation states. At pH < 1, 9d should be completely protonated on the
phosphate group, the amino group on the substituent and N-1.167
Increasing the pH, the
phosphate would lose the first proton (pK1 Pi ≈ 0.9),167
followed by N-1 (pKa N-1 ≈
3.4), the second proton on the phosphate (pK2 Pi ≈ 6.6) and, finally, the primary amine
on the substituent (pKa NH2 ≈ 9.2) (Figure 2.3). These pKa values are comparable with
those in the literature for the related groups.167
Chapter 2
74
Figure 2.3 Acid-base titration of 3-(aminomethyl)phenyl AMP (9d) using a pH meter. To a solution of 9d
in water (pH 6.7), 370 μL of 0.1 N HCl (left) and 390 μL of 0.1 N NaOH (right) were added respectively
in portions of 10 μL each. The pH was read and recorded after each addition. The equivalence point
(E.P.) volumes were found plotting the first derivative [ΔpH/ΔmL] as a function of the titration volume
[μL]: each maximum in the first derivative corresponded to the equivalence point volume. The pKa values
were calculated in correspondence of each half-equivalence point (red circles), where pH = pKa.
When 9d was synthesized by Suzuki-Miyaura cross-coupling of 9a with 3-
(aminomethyl)phenyl boronic acid (Scheme 2.16), two different new spots could be
observed by TLC during the cross-coupling reaction. In this case, it was possible to
isolate by ion-pair purification the two corresponding different products, and investigate
their nature by 1
H NMR and HR-MS. Both techniques confirmed the same identity for
both products as 9d; however, differences in the 1H NMR spectra, in particular in the
phenyl and CH2NH2 regions, suggested their forms as two possible distinct conformers
(Figure 2.4). Examining the CH2NH2 peaks, the enantiotopic protons appeared as a
singlet in the possible conformer a, where the aminomethyl substituent heads far from
the phosphate; indeed, in this case these protons would be too similar to be
distinguished. By contrast, in the other possible conformer b, stabilized by salt
formation between the 5′-phosphate on the adenosine and the amino group on the
substituent, the enantiotopic protons would have a very different electronic environment
and, therefore, appear as a doublet.
3
4
5
6
7
8
9
10
11
-450 -350 -250 -150 -50 50 150 250 350 450
pH
Titration volume (HCl/NaOH) [mL]
pKa N-1
pK2 Pi
pKa NH2
Ka N-1 K2 Pi Ka NH2
E.P.
E.P.
E.P.
Chapter 2
75
Figure 2.4 1H NMR spectra of the two isolated conformers, a and b, for 9d.
To support this hypothesis, it was observed that 9d(a) had 1.3 mol TEA as a counter
ion, while 9d(b) had only 0.5 mol TEA because of the involvement of the phosphate
group in a salt formation with the amino group of the substituent. Hence, it was
suggested that, when 9d was synthesized by deprotection of 9c, only the kinetically
favoured conformer 9d(a) was obtained. Indeed, in 9c the hindered substituent
containing the Boc group would assume the conformation where it heads far from the
phosphate; since the deprotection occurred rapidly, the product 9d would preferentially
remain in the conformation assumed by its precursor. In contrast, when 9d was
synthesized by Suzuki-Miyaura cross-coupling with the unprotected boronic acid, the
substituted phenyl ring was able to rotate and to assume both the different positions
compared to the phosphate group, leading to the formation of both kinetically and
thermodynamically favoured conformers. An useful experiment to prove the existence
of the two different conformers would be to increase the temperature during the 1H
NMR experiment; indeed, the energy provided would allow the phenyl substituent to
rotate freely around its bond with C-8 of the adenine ring, avoiding the stabilization of
the two different conformers and giving an average NMR signal for phenyl and
CH2NH2 peaks.
Phenyla)
b)
CH2NH2
Chapter 2
76
2.6 Synthesis of aryl/heteroaryl adenine-modified NAD+ derivatives (pathway a)
2.6.1 AMP activation via phosphoromorpholidate intermediate
The formation of the pyrophosphate bond between the aryl/heteroaryl adenine-modified
AMP derivatives and β-NMN to generate the final aryl/heteroaryl adenine-modified
NAD+ derivatives, requires the activation of either phosphate group of the reactants.
The Khorana-Moffatt procedure, involving the coupling between a nucleotide and a
nucleotide phosphoromorpholidate, was chosen in our project to perform the
pyrophosphate bond formation. Indeed, the nucleotide phosphoromorpholidate
intermediates are reactive enough phosphate derivatives, but also sufficiently stable to
be handled and stored.168, 169
Because of the fragility of β-NMN due to its labile nicotinamide N-glycosidic bond, it
was chosen to synthesize the phosphoromorpholidate intermediates of the
aryl/heteroaryl adenine-modified AMP derivatives, to subsequently couple them with β-
NMN. The synthesis of the phosphoromorpholidate intermediates was performed by an
oxidation-reduction condensation under the Mukaiyama‟s conditions,170, 171
which had
already been successfully used for the preparation of different pyrophosphate bonds.112,
151 A complete conversion occurred from all the starting materials and the obtained
products were usually precipitated as sodium salts and used in the next step without any
further purification (Scheme 2.17).
Scheme 2.17 Synthesis of aryl/heteroaryl adenine-modified AMP-morpholidate derivatives. (i)
Morpholine, 2,2′-dithiopyridine, PPh3, anhydrous DMSO, rt, 1 h, 70-97% (X is an aryl/heteroaryl group,
see Scheme 2.18, Table 2.2).
Chapter 2
77
2.6.2 Coupling with β-NMN
Once the aryl/heteroaryl adenine-modified AMP-morpholidate derivatives were
synthesized, it was possible to proceed to their coupling with β-NMN to obtain the final
aryl/heteroaryl adenine-modified NAD+ derivatives. As described in the literature,
110 the
Khorana-Moffatt coupling for the synthesis of β-NAD+ needs the presence of Lewis
acid additives to make the morpholidates more reactive. Indeed, β-NMN in the reaction
conditions (pH 7) is probably an inner salt, with a poorly nucleophilic phosphate and,
therefore, unable to efficiently attack the activated phosphoromorpholidates. A MnCl2
solution in formamide has been used for this purpose.110-112, 172
A fundamental
requirement for the success of the reaction is the dryness of the system; indeed, because
of its nucleophilicity, the water could hydrolyse the phosphoromorpholidates or the
newly formed pyrophosphate bond, decreasing the final yields. Therefore, the MnCl2
solution in formamide needs to be dried over molecular sieves for several days prior its
use. Furthermore, anhydrous MgSO4 is added to the reaction mixture to capture any
excess water.
The literature conditions report the use of 1.1 equiv. of β-NMN to the morpholidate.
However, because of the high value of β-NMN, the use of its excess was reconsidered
in our procedure. In contrast, where possible, an excess of morpholidate was used to
improve the reaction yields (Scheme 2.18, Table 2.2). The aryl/heteroaryl adenine-
modified NAD+ derivatives, 13a-b, 14a-b, 15b, c, e, g, h, were obtained with yields
ranging from 24 to 87%. The same variability along the reaction yields was noticed in
the literature, where the Khorana-Moffatt coupling gave the NAD+ derivatives with
yields between 30 and 70%.110-112
Chapter 2
78
Scheme 2.18 Synthesis of aryl/heteroaryl adenine-modified NAD+ derivatives. (ii) β-NMN, anhydrous
MgSO4, 0.2 M MnCl2 in formamide, rt, 12 h, 24-87% (R is an aryl/heteroaryl group, see Table 2.2).
compd Position of substitution R (substituent) Yield (%)a
13a 2 Iodo 53
13b 2 Phenyl 28
14a 6 Phenyl 56
14b 6 Pyrrol-2-yl 87
15b 8 Phenyl 41
15c 8 3-(N-Boc-aminomethyl)phenyl 61
15e 8 Pyridin-3-yl 24
15g 8 2,4-DMT-pyrimidin-5-yl 43
15h 8 Pyrrol-2-yl 30 aYield of isolated products
Table 2.2 Synthesis of the aryl/heteroaryl adenine-modified NAD+ derivatives 13a-b, 14a-b, 15b,c,e,g,h.
Chapter 2
79
β-NAD+ and its derivatives are relatively fragile compounds and need to be handled
extremely carefully to prevent their decomposition. As discussed in 1.1, the
nicotinamide N-glycosidic bond is sensitive to basic conditions, since the hydroxyde
can act as a nucleophile and substitute the nicotinamide ring to give the corresponding
ADPR derivative. During the acquisition of 31
P NMR spectra, the addition of
triethylamine to the samples in D2O is required to sharpen the peaks and show the
coupling constants, JP,P, between the two phosphorus of the pyrophosphate bond.
Therefore, the acquisition of these spectra is particularly tricky for the NAD+
derivatives, since the smallest quantity of TEA has to be added to the cold samples to
avoid their immediate degradation, detected by a sudden change in the colour of the
solution from colourless to orange.
2.7 Alternative synthetic strategies to aryl/heteroaryl adenine-modified NAD+ derivatives
2.7.1 Aryl/heteroaryl adenine-modified NAD+ derivatives via phosphoramidite method (pathway b)
One of the approaches to shorten the synthesis of the NAD+ derivatives is the
phosphoramidite method. The phosphoramidite chemistry is commonly used for the
solid-phase synthesis of oligonucleotides, where a 3′-phosphoramidite nucleoside
intermediate reacts with the 5′-OH of a nucleoside to form the 3′,5′-phosphodiester
bond. The phosphoramidite intermediate is very reactive, allowing successful couplings
with almost quantitative yields, a fundamental requirement for the synthesis of long
oligonucleotide sequences.173-175
In the case of the formation of the pyrophosphate bond
in dinucleotides, this method allows the coupling between a phosphoramidite nucleoside
intermediate and the phosphate group of a nucleotide.113
For the synthesis of aryl/heteroaryl adenine-modified NAD+ derivatives, the substitution
of the adenosine with aryl/heteroaryl groups in the favourite position is followed by the
preparation of the corresponding reactive phosphoramidite intermediate, ready to
undergo the coupling with β-NMN (Scheme 2.19). Therefore, the phosphorylation and
activation of the phosphate occur in a single step and with the purification of a single
intermediate compared to the multistep synthesis (pathway a).
Chapter 2
80
In this project, the pathway b was tested starting from the 2′,3′-O-isopropylidene-2-
phenyl adenosine (4b), forming its corresponding phosphoramidite intermediate and
attempting its coupling with β-NMN under the conditions described in the literature.113
However, the coupling turned out to be unsuccessful despite the high reactivity of the
phosphoramidite intermediate. The reason could be due, as in the Khorana-Moffatt
procedure, to the poor nucleophilicity of the phosphate of β-NMN in the reaction
conditions. In addition, the insolubility of β-NMN in the reaction solvent (acetonitrile)
could negatively influence the effectiveness of the reaction. The addition of small
DMF/DMSO percentages in the reaction media was tried to improve the solubility of β-
NMN; however, the reaction did not proceed. Further investigations and optimizations
studies should be performed to exploit the potential of this synthetic pathway.
Scheme 2.19 Synthesis of 13b via phosphoramidite method (pathway b). (i) Phosphoramidite synthesis:
2-cyanoethyl N,N-diisopropylchlorophosphoramidite, DIPEA, anhydrous CH2Cl2, rt, 1 h; (ii) coupling
with β-NMN: β-NMN, DCI, anhydrous acetonitrile, rt, 1 – 2 h; then, 5.0 – 6.0 M tBuOOH in nonane, rt, 1
h; NH4OH, rt, 3 h; (iii) 2′,3′-O-isopropylidene deprotection: Dowex 88 (H+), rt, 12 h.
2.7.2 Aryl/heteroaryl adenine-modified NAD+ derivatives via one-pot, two-step procedure (pathway c)
β-NAD+ is a very sensitive molecule, difficult to handle and to work with because of the
labile bonds present in its structure. Its derivatives are rarely synthesized directly from
the natural β-NAD+ because of its easy degradation even under mild reaction
conditions. For this reason, the most common pathway to synthesize NAD+ derivatives
is the multistep synthesis (pathway a, Scheme 2.1).
The direct introduction of aryl/heteroaryl substituents on the β-NAD+ scaffold by
Suzuki-Miyaura cross-coupling has been an open challenge, since β-NAD+ is
susceptible to the required basic reaction conditions. The first step for the substitution of
Chapter 2
81
the adenine ring of β-NAD+ by Suzuki-Miyaura cross-coupling is its halogenation in the
favoured position. References for the halogenation of β-NAD+ have been found in the
literature,107, 108
using neat bromine to brominate the reactive C-8 of the adenine ring at
room temperature in 30 min. Unfortunately, the halogenation of the adenine ring in
other positions is not as straightforward as that on C-8 and, therefore, this pathway can
exclusively be applied to the synthesis of 8-aryl/heteroaryl NAD+ derivatives.
The subsequent Suzuki-Miyaura cross-coupling on C-8 requires an accurate
optimization of every reaction parameters, such as the catalyst, the amount of base, the
temperature and the reaction time, to guarantee the occurrence of the cross-coupling
before the beginning of the degradation of the starting material at basic pH.70
As already
shown in the Wagner group, the combination of Na2PdCl4 with the electron-rich
phosphine, TXPTS, allowed fast Suzuki-Miyaura cross-couplings at 40 °C on 8-bromo
NAD+, using aryl/heteroaryl, electron-rich and electron-poor substituents (Scheme
2.20).70
To further shorten this synthesis, a one-pot, two-step procedure was set up, where,
immediately after the bromination, the aqueous solution was extracted with CHCl3 to
remove the bromine excess, and the residue evaporated to dryness. The crude material
was directly used in the Suzuki-Miyaura cross-coupling, adding more equivalents of
base to bring the pH 4 to the optimum pH 9. The yields obtained with this method were
comparable to those obtained with the multistep synthesis.70
Scheme 2.20 Synthesis of 8-aryl/heteroaryl adenine-modified NAD+ derivatives via one-pot, two-step
procedure (pathway c). (i) Bromination: neat bromine, 1 M NaOAc buffer (pH 4), rt, 30 min; then,
Suzuki-Miyaura cross-coupling: aryl/heteroaryl boronic acid, Na2PdCl4, TXPTS, K2CO3 in water, 40 °C,
20 – 30 min, 36-64% (experimental procedures in reference 70).
Chapter 2
82
Although this method works for a large range of boronic acids, including aryl/heteroaryl
and electron-rich/electron-poor ones, the Suzuki-Miyaura cross-couplings with
pyridinyl and pyrrolyl boronic acids required a too basic pH and high temperature to be
applied to the β-NAD+ without any decomposition. The use of surfactants, discussed in
2.5.2, to perform the couplings with electron-poor boronic acids at low temperature was
attempted during the one-pot, two-step procedure with β-NAD+ without any positive
result. Therefore, for certain classes of boronic acids, this pathway still remains
unsuccessful and needs further investigations.
2.8 Conclusions
Owing to the importance of β-NAD+ as a substrate for many different enzymes involved
in major biological processes, the interest toward the synthesis of NAD+ derivatives as
biochemical tools has become more and more relevant, making necessary the
development of efficient, rapid and robust synthetic routes.
After the optimization of the already known multistep synthesis (pathway a), which
allows us to synthesize any NAD+ derivatives starting from the nucleoside precursors,
new routes were explored to improve and shorten the synthesis of NAD+ derivatives.
The phosphoramidite strategy (pathway b) failed but, because of its attractive potential
in remarkably shortening the synthesis, still deserves further investigations. The one-
pot, two-step synthesis (pathway c), although useful exclusively for the synthesis of 8-
aryl/heteroaryl NAD+ derivatives, has successfully allowed the synthesis of several
NAD+ derivatives in a very short time.
Therefore, a small library of NAD+ derivatives substituted on C-2, C-6 or C-8 of the
adenine ring with aryl/heteroaryl groups was generated, to be tested on different NAD+-
consuming enzymes as inhibitors for drug discovery or NNSs as biochemical tools.
Chapter 3
83
3 CONFORMATIONAL AND SPECTROSCOPIC ANALYSIS OF ARYL/HETEROARYL ADENINE-MODIFIED AMP AND NAD+ DERIVATIVES
The NAD+ derivatives synthesized in Chapter 2 were designed for their biological
application as inhibitors and/or NNSs toward NAD+-consuming enzymes. To perform
their functions, the compounds have to interact with the enzymes and their active sites,
consequently requiring particular conformations that are adaptable to the enzyme
recognition. A certain degree of flexibility in the molecule is required for enzyme
recognition. In particular, in the case of nucleosides/nucleotides, a fundamental feature
is the orientation of the nucleobase in relation to the ribose. Of course, the presence of
additional groups on the adenine-modified NAD+ derivatives, able to change the natural
nucleobase orientation and the normal H-bonding or hydrophobic interactions of the
natural β-NAD+ within the enzyme active sites, influences the recognition and its
strength, and it is responsible for the compounds activity as inhibitors and/or NNSs.
In view of the potential activity of aryl/heteroaryl adenine-modified NAD+ derivatives
as NNSs for NAD+-consuming enzymes, their spectroscopic features could lead to their
application as biochemical probes for the development of sensitive assays based on
spectroscopic techniques, such as UV/Vis absorption or fluorescence emission
spectroscopies. For example, it is known, as will be discussed in 3.3.2, that adenine
rings, substituted in position 8 with aryl/heteroaryl groups, belong to the class of
emissive nucleobases.
Therefore, this chapter will first describe the conformational properties of the natural
AMP and β-NAD+, and their derivatives synthesized in Chapter 2, to help later correlate
structures with the bio-activities observed. Then, their spectroscopic features will be
described to explore the suitability of aryl/heteroaryl NAD+ derivatives as biochemical
probes for NAD+-consuming enzymes.
Chapter 3
84
3.1 Conformational analysis of nucleosides and nucleotides
Nucleosides and nucleotides are characterized by free or restricted rotation around
several bonds, which allows them to assume different conformations in solution. The
topology of a nucleoside in solution is determined by interdependent conformational
equilibria involving the most flexible bonds in the structure. The nucleobase is oriented
perpendicularly to the plane of the ribose ring and can mainly assume two different
conformations, syn and anti, by rotation around the N-glycosidic bond. The torsion
angle around this bond, χ, determines which conformation is favoured (Figure 3.1).176
Figure 3.1 Glycosidic conformational ranges of nucleosides and nucleotides.
The rotation around the N-glycosidic bond is not free, due to the steric hindrance
between the nucleobase and the ribose. For this reason, the anti conformation, in which
the nucleobase is pointing away from the ribose, is generally favoured by
nucleosides/nucleotides compared to the syn conformation, which has a higher energy
(Figure 3.2).
270°
180°
90°
0°
345°
165°
75°
255°
High syn syn
O1
C2
C3
C4
C1
Planeof base
High anti
antiχ
Chapter 3
85
Figure 3.2 Adenine conformations in the natural adenosine/AMP and the aryl/heteroaryl adenine-
modified adenosine/AMP derivatives.
The syn/anti equilibrium largely influences the ribose conformation: indeed, the ribose
can pucker its structure, assuming the North (N, or C3′-endo) or South (S, or C2′-endo)
conformation (Figure 3.3), in equilibrium to each other. The general ratio between N
and S conformations for β-D-ribonucleosides is 1:1, but because the N-conformation is
more hindered than the S-conformation, nucleosides/nucleotides with the nucleobase in
the syn orientation will prefer the S to the N conformation.167, 177
Figure 3.3 Ribose conformations of nucleosides (R = H) and nucleotides (R = Pi).
Chapter 3
86
3.1.1 Conformation of aryl/heteroaryl adenine-modified AMP derivatives
A conformational analysis was performed on the various substituted AMP derivatives
synthesized in Chapter 2 (Table 3.1). The natural AMP exists predominantly with the
adenine in anti conformation, while the less hindered S-conformation is slightly
favoured by the ribose.178
To determine the preference of the adenine between syn and
anti conformation in the AMP derivatives, it is possible to observe the difference in
chemical shift of H-2′ between the natural AMP and the different derivatives (ΔH-2′). In
particular, a downfield shift for H-2′ compared to its chemical shift in AMP indicates a
deshielding due to the adenine assuming a syn conformation and, consequently, the
ribose favouring an S-conformation. The coupling constant, J1ꞌ,2ꞌ, is related to the
percentage of the ribose population in S conformation.179, 180
cmpd Position of
substitution R (substituent)
S-type conformer
(%) (10J1ꞌ,2ꞌ)
H-2′
(ppm)
ΔH-2′
(ppm)a syn/anti
AMP H 52 4.75 – anti
7a 2 Iodo 57 4.72 -0.03 anti
7b 2 Phenyl 49 4.68 -0.07 anti
8a 6 Phenyl 53 4.68 -0.07 anti
8b 6 Pyrrol-2-yl 55 4.75 0 anti
9a 8 Bromo 60 5.19 0.44 syn
9b 8 Phenyl 62 5.21 0.46 syn
9c 8 3-(N-Boc-
aminomethyl)phenylb 58 5.51 – syn
9d 8 3-(Aminomethyl)phenyl 62 a); 64 b) 5.17 0.42 syn
9e 8 Pyridin-3-y – 5.22 0.47 syn
9f 8 Pyridin-4-yl 51 5.27 0.52 syn
9g 8 2,4-DMT-pyrimidin-5-yl 62 5.19 0.44 syn
9h 8 Pyrrol-2-yl 62 5.19 0.44 syn
9i 8 Furan-2-yl 57 5.31 0.56 syn
9j 8 1H-Pyrazol-4-yl 62 5.25 0.50 syn
9k 8 5-Formyl-thien-2-yl – 5.30 0.55 syn
9l 8 5-Bromo-thien-2-yl 61 5.42 0.67 syn
9m 8 Indol-2-yl 57 5.42 0.67 syn
9n 8 Benzofuran-2-yl 53 5.39 0.64 syn aΔH-2′ has been calculated as the difference between the chemical shift of H-2′ of the aryl/heteroaryl
adenine-modified AMP derivatives and H-2′ of AMP (4.75 ppm) in D2O (referenced for D2O, 4.79
ppm). b 1
H NMR spectrum for 9c recorded in CD3OD.
Table 3.1 Conformational analysis of the various aryl/heteroaryl adenine-modified AMP derivatives
synthesized in this study.
Chapter 3
87
As shown in Table 3.1, a clear trend amongst the various substituted AMP derivatives
could be observed. In particular, AMP derivatives substituted on C-2 and C-6 adopted a
favoured anti conformation and an S-type conformer percentage ≥ 50. In contrast, AMP
derivatives substituted on C-8 preferred a syn conformation with an S-type conformer
percentage > 50. Looking at the molecular structures of the several AMP derivatives
(Figure 3.2), these were the expected behaviours. Indeed, AMP derivatives substituted
on C-2 were very hindered on the pyrimidine ring and, therefore, they could not afford a
syn conformation which would bring a huge steric hindrance with the ribose. Regarding
the AMP derivatives substituted on C-6, that have the substituent heading in the
opposite direction compared to the ribose, they could assume the same anti
conformation as the natural AMP. Finally, AMP derivatives substituted on C-8
preferred to place the substituents far from the ribose, assuming therefore a syn
conformation, in which the pyrimidine ring steric hindrance was less strong than that
deriving from the substituents.
3.1.2 Conformation of aryl/heteroaryl adenine-modified NAD+ derivatives
The conformations of the NAD+ derivatives followed approximatively the same trend
described for the corresponding AMP derivatives. Therefore, the NAD+ derivatives
substituted on C-2 had the adenine in anti conformation, slightly turning into syn
conformation compared to the favoured anti conformation in the AMP derivatives. The
NAD+ derivatives substituted on C-8 generally preferred the adenine in syn
conformation and the ribose in S (Table 3.2).70
Slight differences of S-type conformer
percentage and ΔH-2′ in the NAD+ derivatives compared to the AMP derivatives, were
due to a different arrangement of the riboses and the adenine ring within the β-NAD+
structure. Indeed, β-NAD+ is a very flexible molecule because of a variety of rotatable
bonds and many different functional groups, which are able to interact in several ways
with each other, with the solvent and/or with the amino acids of the enzymatic pockets
where β-NAD+ is the substrate.
Chapter 3
88
cmpd Position of
substitution R (substituent)
S-type conformer
(%) (10J1ꞌ,2ꞌ)
H-2′
(ppm)
ΔH-2′
(ppm)a syn/anti
β-NAD+
- H 55 4.71 – anti
13a 2 Iodo 53 4.74 0.03 anti
13b 2 Phenyl – 4.84 0.13 anti
15b 8 Phenyl 59 5.14 0.43 syn
15c 8 3-(N-Boc-
aminomethyl)phenyl – 5.16 0.45 syn
15e 8 Pyridin-3-yl 57 5.24 0.53 syn
15g 8 2,4-DMT-pyrimidin-5-yl 55 5.08 0.37 syn
15h 8 Pyrrol-2-yl 48 5.33 0.62 syn aΔH-2′ has been calculated as the difference between the chemical shift of H-2′ of the aryl/heteroaryl
adenine-modified NAD+ derivatives and H-2′ of β-NAD
+ (4.71 ppm) in D2O (referenced for D2O,
4.79 ppm).
Table 3.2 Conformational analysis of the various aryl/heteroaryl adenine-modified NAD+ derivatives
synthesized in this study.
3.2 Nucleobases and their photochemical properties
Absorption and emission spectroscopies are based respectively on the measurement of
light absorbed or emitted by molecules. Depending on the energy of the absorbed light,
different phenomena can occur in the molecules and, consequently, different techniques
have been developed to explore them.
UV/Vis absorption spectroscopy looks at the absorption by the molecules of UV/Vis
light, which promotes the electrons in the external shell from the ground state, S0, to an
electronically excited state, S1. Then, the excited electrons decay back to the S0
dissipating the energy by different processes, non radiative in the form of heat, or
radiative by emission of photons. The last process is called photoluminescence.
Photoluminescence processes can be divided into fluorescence (F) and phosphorescence
(P) (Figure 3.4). The former occurs when excited and ground states have the same spin
multiplicities (S1 S0); by contrast, the latter occurs when they have different spin
multiplicities (T1 S0). The quantum yield (Φ) is a relative fluorescence measurement
of the compounds, indicating the fraction of excited molecules that decay back by
fluorescence emission.181-183
Chapter 3
89
Figure 3.4 Physical processes occurring when a molecule absorbs UV/Vis light. S0 is the electronic
ground state of the molecule; S1 and T1 are respectively the excited singlet and triplet states at lower
energy. The straight arrows indicate radiative processes with emission of photons,while the curly ones the
non radiative processes to dissipate energy. A (Absorption), F (Fluorescence), P (Phosphorescence), IC
(Internal Conversion), ISC (Inter System Crossing).
Fluorescence spectroscopy is a very sensitive technique, largely used in biochemistry to
investigate and understand many major biochemical topics (e.g. structure, binding,
recognition). Intrinsically fluorescent biological molecules are present in nature and act
as natural biochemical probes (e.g. β-NADH, tryptophan, phenylalanine and tyrosine).85
In the case of nucleic acids, the most common nucleobases, such as purines and
pyrimidines, are generally not emissive in neutral aqueous conditions. However, rare
exceptions of naturally occurring emissive nucleobases can be found in nature and used
as “built-in” probes.47, 184, 185
In particular, adenine and its corresponding nucleoside, adenosine, can be considered
not emissive, since their quantum yields are very low (respectively 3 x 10-4
and 5 x 10-5
)
and the excited-state lifetimes of the order of picoseconds. In adenine, the decay to the
ground state occurs by internal conversion, adopted by the nucleobases as a natural
defence to protect the nucleic acids from the photodecomposition. Surprisingly, the
adenine isomer, 2-amino purine (2-AP), where the NH2 is in position 2 instead of 6 on
the purine ring, is strongly emissive (Φ = 0.66, and 0.68 for the ribonucleoside) with a
long excited-state lifetime. In this case, the electronic stabilization of the C-2
surrounded by three nitrogens would decrease the energy of the excited state,
minimizing the internal conversion of the excited molecules to the ground state. This is
just an example of how small changes in the nucleobase structure can modify
completely its photochemical properties.186-189
S0
S1
T1
A FP
IC ISC
Chapter 3
90
Owing to the importance of the nucleic acids and their crucial role in life, a deep
knowledge about their structure and the biological processes in which they are involved,
such as DNA replication, repair, and recombination, transcription and translation, and
many others, are some of the most attractive, challenging and always relevant source of
scientific interest. In addition, nucleosides, nucleotides and their derivatives are
involved as substrates or signalling molecules in many cellular processes. Therefore, the
study of their activities and binding to the proteins is important to understand
fundamental biological processes. For this reason, major efforts have been undertaken
to synthesize emissive nucleobases that, introduced in nucleosides and their derivatives,
allow their study by fluorescence spectroscopy.
3.3 Design of emissive nucleosides
3.3.1 Fluorophore features
For the design of emissive nucleosides, many factors influencing the fluorescence
properties have to be taken in consideration. The main features that characterize a good
fluorophore are summarised below:
Extended electron delocalization; the more extended the delocalization, the more
stable and at lower energy (red shift) is the excited state, minimising the
photodecomposition of the molecule.
Rigidity of the fluorophore; too many rotational and vibrational levels in the excited
state can cause the decay to the ground state via vibrational relaxation, decreasing
the fluorescence emission.
Quantum yield of the fluorophore (Φf) is the ratio of photons absorbed to photons
emitted by fluorescence; the maximum value for Φf is 1. The higher the quantum
yield, the stronger is the fluorescence of the fluorophore and, consequently, the
sensitivity of the fluorescence detection, even at low concentrations of the
fluorophore.
Chapter 3
91
Solvatochromism, or fluorescence changes of the molecules depending on the
surrounding environment, such as the polarity and the H-bonding ability of the
solvent. The solvent molecules that solvate the fluorophore assist the stabilization of
its excited state by solvent relaxation. The stronger the solvation, the higher is the
solvent relaxation; hence, the more stable and at lower energy (red shift) is the
excited state and the larger is the distance between λex and λem (Stokes shift).
Quenching, or any process which decreases the fluorescence of a molecule. It can be
caused by quenchers present in the sample, such as O2, or by the same fluorophore,
if present in too high concentrations (self-quenching). The quenching can be due to
intra/intermolecular collisions (dynamic quenching); or to the formation of not
emissive fluorophore-quencher complexes (static quenching).181-183
Keeping in mind all these factors influencing the fluorescence, it is possible to design
fluorophores with suitable features to their use.
3.3.2 Families of existing nucleobase-modified emissive nucleosides
Several approaches have been followed to obtain emissive nucleosides, such as the
modification on the nucleobase or the ribose.47, 185
Herein, the attention has been
focused on the main families of emissive nucleosides modified on the nucleobase.184, 190
Figure 3.5 shows the emissive nucleosides obtained by the replacement of the
nucleobases with well known fluorophores, as polycyclic aromatic hydrocarbons
(PAH), or natural heterocycles, as pteridines; or by the introduction on the nucleobases
of fluorophores not directly conjugated with them. This family of emissive nucleosides
is characterized by the fluorescent properties of the attached fluorophores.191
Since the fluorescence is due to an extended delocalization of π-electrons, it is possible
to obtain emissive nucleosides extending the electronic delocalization of the natural
nucleobases. From this approach, the so-called expanded and extended nucleobases
have been generated (Figure 3.6).
Chapter 3
92
Figure 3.5 Emissive nucleosides obtained by replacement of the nucleobases or introduction on their
skeleton of known fluorophores.
Figure 3.6 Expanded and extended nucleosides, obtained by extension of the π-electrons delocalization of
the nucleobases.
Chapter 3
93
The expanded nucleobases have additional aromatic rings fused to the natural
nucleobases to extend their electronic conjugation (inspired by the rare natural wye
nucleosides found in tRNA192, 193
).83, 194-196
In contrast, the extended nucleobases have
known fluorophores connected via electronically conjugated linkers.130, 197-200
The
emissive nucleosides belonging to these families are characterized by peculiar
fluorescent properties, since all the conjugated groups of the molecule are an integral
part of fluorophore.
Finally, the isomorphic nucleobases are heterocycles resembling the natural
nucleobases in size and H-bonding patterns, but with characteristic fluorescent features
(Figure 3.7).151, 152, 172, 200-203
Figure 3.7 Emissive isomorphic nucleosides.
Many of them contain aryl/heteroaryl groups as substituents on the purine and
pyrimidine skeleton. It has already been reported that adenosines substituted on C-8 of
the adenine ring with aryl/heteroaryl groups belong to the class of emissive isomorphic
nucleosides,130, 200, 202
while those substituted on C-2 and C-6 are weakly emissive, as
experimental evidence will suggest in this study.
In this chapter, the spectroscopic analysis of aryl/heteroaryl adenine-modified AMP and
corresponding NAD+ derivatives, synthesized in Chapter 2, will be performed in view
of their possible biochemical applications.
Chapter 3
94
3.4 Measurements of quantum yield for aryl/heteroaryl adenine-modified AMP and NAD+ derivatives
The most common method to measure the unknown quantum yield of a fluorophore, Φx,
is the comparative method,204, 205
which involves the use of standards with known Φst
values. Indeed, assuming that solutions of standard and sample at the same λex absorb
the same number of photons, the ratio between the integrated fluorescence intensities of
the two solutions will be the ratio of their Φ values, from which Φx can be easily
calculated.
In practice, the quantum yield measurements can be affected by many factors that
influence the experimental results. The choice of the standards, which must have similar
λex and, possibly, also λem to the compound with unknown Φx, is critical. The cross-
calibration of two standards with each other is crucial for robust and reliable results,
since the Φst is used to calculate Φx. Only when the cross-calibration gives Φst values in
agreement with those in the literature, it is possible to proceed to the measurement of Φx
with a certain level of confidence. In addition, the choice of a proper range of
concentrations to guarantee a linear response in the absorbance and fluorescence
measurements is fundamental to correctly measure Φx. Indeed, at high concentrations
the self-quenching phenomena are quite common due to intermolecular collisional
energy transfers.
For the Φx measurement, absorbance and fluorescence spectra of sample and standard at
different dilutions are recorded and their integrated area plotted, fluorescence versus
absorbance. The gradient of the two extrapolated lines (Gradx and Gradst) can be
substituted in the following equation [1] and corrected for the refractive indexes of
sample and standard solutions (ηx and ηst), allowing the calculation of Φx:
Φx = Φst (
) (
) [1]
values were calculated for different AMP and NAD+ derivatives synthesized in
Chapter 2, using 2-amino pyridine and tryptophan as standards, previously cross-
calibrated to each other. AMP and NAD+ derivatives substituted respectively in 2 and 6
position on the adenine ring turned out to be poorly emissive. A simple evaluation of
Chapter 3
95
their fluorescence was performed by visualization under UV light (254/365 nm) of their
spots in TLC, showing no trace of fluorescence; values were calculated only for the
6-(pyrrol-2-yl) AMP (8b) and the corresponding NAD+ derivative (14b), which showed
in these conditions to be slightly more fluorescent than the others. In contrast, 8-
substituted AMP and NAD+ derivatives were good fluorophores and, therefore,
values were calculated for many of them.
3.5 Spectroscopic analysis of aryl/heteroaryl adenine-modified AMP derivatives
Table 3.3 summarises the spectroscopic analysis for aryl/heteroaryl adenine-modified
AMP derivatives synthesized in Chapter 2. Different trends in λex, λem, Stokes shift and
values were observed in the series of fluorescent 8-aryl/heteroaryl AMP derivatives
depending on the nature of the substituent; the values followed approximately the same
trends in water and methanol.
cmpd Position of
substitution R (substituent) solvent λex/nm λem/nm
Stokes
shifta/cm
-1
AMP – H Water 260 380 12146 –
7b 2 Phenyl Water 270 374 10299 –
8a 6 Phenyl Water 288 387 8882 –
8b 6 Pyrrol-2-yl Water 340 382 3234 0.0041
9b 8 Phenyl Water 274 381 10250 0.669
MeOH 281 376 8991 0.225
9c 8 3-(N-Boc-
aminomethyl)phenyl
Water 274 383 10387 0.888
MeOH 281 378 9132 0.639
9e 8 Pyridin-3-yl Water 277 401 11163 0.0034
MeOH 287 406 10213 0.021
9f 8 Pyridin-4-yl Water 284 421 11458 0.032
MeOH 293 422 10433 0.044
9g 8 2,4-DMT-pyrimidin-5-yl Water 270 410 12647 0.021
MeOH 272 404 12012 0.147
9h 8 Pyrrol-2-yl Water 306 374 5942 0.231
MeOH 306 376 6084 0.272
9i 8 Furan-2-yl Water 302 378 6658 0.538
MeOH 300 375 6667 0.638
9k 8 5-Formyl-thien-2-yl Water 334 436 7004 0.084 a Stokes shift = (1/ex - 1/em).
Table 3.3 Spectroscopic analysis in water and methanol of natural AMP and aryl-heteroaryl adenine-
modified AMP derivatives synthesized in this study.
Chapter 3
96
Firstly, the absorbance spectra of 8-aryl/heteroaryl AMP derivatives were studied. They
presented two absorbance bands due to different electronic transitions, that at longer
wavelengths (lowest energy) diagnostic for the substituents introduced on the adenine
ring and, therefore, for larger π-electron delocalizations. In Figure 3.8, the absorbance
spectra of phenyl (9b), pyridin-3-yl (9e) and pyrrol-2-yl (9h) AMP derivatives are
shown to compare the effect of phenyl, electron-poor and electron-rich substituents on
the λex maxima. Analysing the shift of λex with the substituents, a red shift
(bathochromic effect) occurred moving from the phenyl, through the electron-poor, to
the electron-rich substituted adenine rings (Figure 3.8, Figure 3.9). Indeed, 2,4-DMT-
pyrimidine has the excited state with the highest energy because of the partial aromatic
delocalization due to its tautomeric form. To follow, the phenyl ring with an equal π-
electron delocalization. Then, the pyridine with the energetic difference between the
ground (HOMO) and the excited state (LUMO) smaller than benzene, because of the
non-bonding orbitals containing the nitrogen lone pair (HOMO) at higher energy.
Finally, the electron-rich heterocycles, better π-donors than the six-membered rings,
with the smaller difference between HOMO and LUMO due to the large π-electron
delocalization.
Figure 3.8 Absorbance spectra of AMP and 9b,e,h in water.
0
0.2
0.4
0.6
0.8
1
1.2
200 250 300 350 400
Ab
s
wavelength [nm]
AMP
Phenyl (9b)
Pyridin-3-yl (9e)
Pyrrol-2-yl (9h)
λex
Chapter 3
97
Figure 3.9 Variations of λex with the different substituents in water and methanol for the 8-aryl/heteroaryl
adenine-modified AMP derivatives synthesized in this study.
The fluorescence spectra of 8-aryl/heteroaryl AMP derivatives showed a main emission
band between 370 and 440 nm, typical for the emission of the substituted adenine rings.
A less intense band between 650 and 900 nm could be due to Pd(II) impurities, deriving
from the Suzuki-Miyaura cross-coupling. Indeed, Pd(II) can form Pd(II)-adenine
complexes with the exocyclic NH2 and N-7, or N-1 (Figure 3.10).206, 207
Figure 3.10 Fluorescence emission spectra of 9b,e,h in water at λex maxima (see Table 3.3).
The λem for the 8-aryl/heteroaryl AMP derivatives followed a less clear trend compared
to λex along the different substituents. However, with few exceptions, the trend for λem
was the same followed by the Stokes shift values, that clearly increased from the
electron-rich, through the phenyl, to the electron-poor substituted adenine rings (Figure
3.11). The λem and Stokes shift depend on vibrational and solvent relaxation; the
stronger these two effects, the larger is the Stokes shift and the longer is the λem.
λex [nm]
0
100
200
300
400
500
600
700
800
900
200 300 400 500 600 700 800
Flu
ore
scen
ce E
mis
sio
n
wavelength [nm]
Phenyl (9b)
Pyridin-3-yl (9e)
Pyrrol-2-yl (9h)
Chapter 3
98
Electron-rich substituted adenine rings, having a large π-electron delocalization, are
more rigid and planar than those phenyl and electron-poor substituted, having minor π-
electron distribution. Consequently, λem and Stokes shift, depending on vibrational
relaxation, were larger for electron-poor and phenyl substituted adenine rings, having
many vibrational energetic levels in the excited state, than those that were electron-rich
substituted (Figure 3.11). In addition, electron-poor substituted adenine rings had longer
λem and larger Stokes shift than those phenyl substituted because of a higher solvent
relaxation, due to their H-bonding ability and large dipole moment in the excited state.
The Stokes shift values were generally larger in water than in methanol because of more
H-bonds, which caused a higher solvent relaxation. However, electron-rich substituted
adenine rings appeared unaffected by the solvent change, with their Stokes shift values
unchanged between water and methanol. Therefore, these substituents could be
insensitive to the surrounding environment, as confirmed later from their unchanged
values.
Figure 3.11 Variations of λem, Stokes shift and quantum yield () values with the different substituents in
water and methanol for the 8-aryl/heteroaryl adenine-modified AMP derivatives synthesized in this study.
λem [nm]
StokesShift [cm-1]
Chapter 3
99
The values for the substituted adenine rings seemed to follow a precise trend along
the introduced substituents, as well as λex, λem and Stokes shift. In particular, they
increased moving from the electron-poor, through the electron-rich, to the phenyl
substituted adenine rings (Figure 3.11). The trend was similar between water and
methanol. From the previous considerations, electron-rich substituted adenine rings
would be expected to be the most fluorescent in the series because of a larger π-electron
delocalization, rigidity and planarity, and limited sensitivity to the solvent effect.
However, a new factor, the dynamic/static quenching, could influence values. Indeed,
the strong planarity of electron-rich substituted adenine rings would favour the
intermolecular base stacking interactions with consequent fluorescence decrease.
The values should increase in methanol compared to water because of a minor
solvent relaxation. This effect was observed for electron-poor substituted adenine rings;
in contrast, values decreased for those phenyl substituted, and remained
approximately the same for electron-rich substituted adenine rings. In the first case, the
strongly emissive phenyl substituted adenine rings could undergo a dynamic/static
fluorescence quenching, decreasing overall their values. Regarding the electron-rich
substituted adenine rings, the unchanged values could be due to their insensitivity to
the solvent, already observed in the Stokes shift values.
The spectral properties of AMP derivatives with the same substituent but in different
positions were analysed (Table 3.4, Figure 3.12). Moving from the AMP derivatives
substituted on C-6 to those substituted on C-2 and C-8, λex, even more than λem,
underwent a substantial blue shift (hypsochromic effect), responsible for an increase of
Stokes shift. In addition, in the case of 8-substituted AMP derivatives, a significant
increase in was observed. Therefore, it was suggested that the substitution of the
exocyclic amino group in position 6 on the adenine ring with aryl/heteroaryl groups
helped stabilize the electronic configuration of the 6-substituted adenine rings,
decreasing the gap between HOMO and LUMO, larger otherwise in 2 and 8-substituted
adenine rings. In addition, the removal of the exocyclic amino group would decrease the
H-bonding ability of 6-substituted AMP derivatives and, consequently, their Stokes
shift values.
Chapter 3
100
Phenyl substituent Pyrrol-2-yl substituent
cmpd C λex/nm λem/nm Stokes
shifta/cm
-1
cmpd C λex/nm λem/nm Stokes
shifta/cm
-1
7b 2 270 374 10299 – – – – – – –
8a 6 288 387 8882 – 8b 6 340 382 3234 0.0041
9b 8 274 381 10250 0.669 9h 8 306 374 5942 0.231 a Stokes shift = (1/ex - 1/em).
Table 3.4 Comparative spectroscopic analysis of adenine-modified AMP derivatives with the same
substituent (phenyl and pyrrol-2-yl) in different positions.
Figure 3.12 Comparison of fluorescence of adenine-modified AMP derivatives with the same/similar
aryl/heteroaryl substituent in different positions. To notice the influence of the position of the exocyclic
amino group on the fluorescence. The direction of the arrow indicates an increase in fluorescence.
It is important to notice the change in fluorescence related to the position of the
substituent and the exocyclic amino group on the purine ring (Table 3.4, Figure 3.12).
As discussed in 3.2, 2-AP is more fluorescent than adenine because of the presence of
the exocyclic amino group on C-2 instead of C-6. This relatively simple change of
Chapter 3
101
position of the substituent decreases the internal conversion of the excited molecules to
the ground state and, consequently, increases their fluorescence. In the same way,
looking in Figure 3.12 at the differently substituted adenine rings, it can be observed
that, when the substituent is on the pyrimidine ring, as in 2-amino-6-(2-thienyl)
purine,203
8a, 8b and 7b, the presence of the exocyclic amino group in the specific
position 2 is fundamental to generate the fluorescence, more than its absence/presence
in generic positions. In contrast, its presence in position 6 does generate fluorescence
only if the substituent is on the imidazole ring. Therefore, energetic differences in
excited states of adenine rings substituted in different positions are responsible for their
quantum yield changes.
3.6 Spectroscopic analysis of aryl/heteroaryl adenine-modified NAD+ derivatives
Table 3.5 summarises the spectroscopic analysis for aryl/heteroaryl adenine-modified
NAD+ derivatives synthesized in this study. As for the AMP derivatives, different trends
in λex, λem, Stokes shift and values were observed in the series of fluorescent 8-
aryl/heteroaryl NAD+ derivatives depending on the nature of the substituents.
cmpd Position of
substitution R (substituent) solvent λex/nm λem/nm
Stokes
shifta/cm
-1
β-NAD+ – H Water 259 380 12294 –
13b 2 Phenyl Water 268 380 10998 –
14a 6 Phenyl Water 289 400 9602 –
14b 6 Pyrrol-2-yl Water 344 388 3297 0.0012
15b 8 Phenyl Water 272 381 10518 0.2067
MeOH 274 378 10041 0.080
15c 8 3-(N-Boc-
aminomethyl)phenyl
Water 274 383 10387 0.150
MeOH 273 379 10245 0.148
15e 8 Pyridin-3-yl Water 273 401 11692 0.0035
MeOH 276 399 11169 0.016
15g 8 2,4-DMT-pyrimidin-5-yl Water 268 412 13042 0.0026
MeOH 267 411 13122 0.0054
15h 8 Pyrrol-2-yl Water 309 374 5624 0.0053
MeOH 307 371 5619 0.055
a Stokes shift = (1/ex - 1/em).
Table 3.5 Spectroscopic analysis in water and methanol of natural β-NAD+ and aryl-heteroaryl adenine-
modified NAD+ derivatives synthesized in this study.
Chapter 3
102
The trends along the series of the NAD+ derivatives were generally similar to those
observed for the AMP derivatives. In contrast to the AMP derivatives, the Stokes shift
values for the NAD+ derivatives were approximately the same in water and methanol
(Table 3.5). This suggested that the NAD+ derivatives were generally less influenced by
the solvatochromism effect than the corresponding AMP derivatives. In addition,
values remained generally the same, or increased in methanol compared to water. This
trend, not apparently due to the solvatochromism effect, was attributed to a new factor
occurring in the NAD+ derivatives, such as a dynamic/static fluorescence quenching due
to the intramolecular stacking interactions, or close proximity between the nicotinamide
and the substituted adenine rings. These interactions could be disrupted by methanol or
less polar solvents, which would help β-NAD+ passing from a folded to an extended
conformation. This phenomenon and other evidence for its occurrence will be largely
discussed in 3.7.
Looking carefully at the absorbance spectra of the NAD+ derivatives, it was observed
that the lowest energy band was the result of overlapping bands, due to the substituted
adenine and the nicotinamide ring (Figure 3.13).
Figure 3.13 Absorbance spectrum of 15e in water. The absorbance spectrum is the pink line; the broken
lines (blue) indicate the overlap for the bands related to the substituted adenine and nicotinamide rings.
0
0.2
0.4
0.6
0.8
1
1.2
200 250 300 350 400
Ab
s
wavelength [nm]
Pyridin-3-yl (15e)
Chapter 3
103
The fluorescence emission spectra of 8-aryl/heteroaryl NAD+ derivatives, shown in
Figure 3.14, again presented the two typical bands observed for the AMP derivatives.
Figure 3.14 Fluorescence emission spectra for 15b,c,e,g,h (0.8 μM in water) at λex maxima (see Table
3.5).
Further spectroscopic analyses were performed on 8-aryl/heteroaryl NAD+ derivatives.
In particular, the fluorescence changes with the pH were investigated. Indeed, the
fluorescence can be strongly influenced by the pH, since different protonation states can
generate in the fluorophore different π-electron delocalization and distribution. In the
case of the 8-aryl/heteroaryl NAD+ derivatives, the fluorophore examined was the
substituted adenine ring, which could be protonated on the basic N-1 (Figure 3.15).
Figure 3.15 Protonation of N-1 of 8-aryl/heteroaryl adenine-modified NAD+ derivatives.
0
50
100
150
200
250
200 300 400 500 600 700 800 900
Flu
ore
sc
en
ce
Em
iss
ion
wavelength [nm]
Phenyl (15b)
3-(Boc-aminomethyl)phenyl (15c)
Pyridin-3-yl (15e)
2,4-DMT-pyrimidin-5-yl (15g)
Pyrrol-2-yl (15h)
Chapter 3
104
At acidic pH, the positive charge on the protonated N-1 destabilized the π-electron
delocalization on the adenine ring decreasing its fluorescence (hypochromic effect). By
contrast, at basic pH, the aromaticity of the adenine ring was re-established and the
fluorescence increased (hyperchromic effect).
This behaviour was observed for all 8-aryl/heteroaryl NAD+ derivatives, with the
exception of 15g, where fluorescence decreased as the pH increased (Figure 3.16).
Indeed, at acidic pH, the nitrogens on the 2,4-DMT-pyrimidine would be protonated
with two partial positive charges; therefore, they would be able to form an extended
network of H-bonds, making the substituted adenine ring more rigid and, consequently,
more fluorescent than at basic pH (Figure 3.17). Plotting the fluorescence versus the
pH, it was possible to extrapolate the pKa values for N-1 of 8-aryl/heteroaryl NAD+
derivatives (Figure 3.16).172
Figure 3.16 Fluorescence emission at different pH for 15b,c,e,g,h. Conditions: NaOAc/AcOH buffer (50
mM) was used for pH 3 and 4; KH2PO4/NaOH buffer (50 mM) was used for pH 6; Tris/HCl buffer (50
mM) was used for pH 7, 8 and 9. The compound concentrations were respectively: 1.01 μM for 15b, 3.9
μM for 15c, 3.6 μM for 15e, 3.5 μM for 15g, 0.81 μM for 15h. Fluorescence emission spectra were
recorded at λex maxima (see Table 3.5), rt. Emission spectra areas were normalized to the highest value
for each compound and the values were fitted to the pKa single equation using GraFit5 (except for 15g,
fitted to a single exponential equation).
pH
0 2 4 6 8 10
No
rmaliz
ed E
mis
sio
n S
pectr
a A
rea
0
0.2
0.4
0.6
0.8
1 Phenyl
3-(Boc-aminomethyl)phenyl
3-Pyridinyl
2,4-DMT-pyrimidinyl
2-Pyrrolyl
Phenyl (15b)
3-(Boc-aminomethyl)phenyl (15c)
Pyridin-3-yl (15e)
2,4-DMT-pyrimidin-5-yl (15g)
Pyrrol-2-yl (15h)
Chapter 3
105
The pKa calculated by potentiometric titration for β-NAD+ is 3.88 ± 0.02.
186, 208 As
shown in Table 3.6, the pKa values obtained for 8-aryl/heteroaryl NAD+ derivatives
were close to that of the natural β-NAD+, although following a certain trend depending
on the nature of the substituent, which influenced the basicity of N-1.
cmpd R (substituent on C-8) pKa
β-NAD+ 208
H 3.88 ± 0.02
15b Phenyl 3.81 ± 0.04
15c 3-(N-Boc-aminomethyl)phenyl 3.95 ± 0.14
15e Pyridin-3-yl 2.69 ± 0.18
15h Pyrrol-2-yl 4.07 ± 0.21
Table 3.6 pKa values for the natural β-NAD+ and 15b,c,e,h.
The phenyl substituents in 15b and 15c did not change substantially the π-electron
delocalization of the adenine ring, which had very similar pKa values to β-NAD+. The
electron-poor substituents, withdrawing electrons from the adenine ring, decreased the
pKa of N-1; in particular, the pyridine ring in 15e at acidic pH would be present as the
pyridinium ion, a very strong electron-withdrawing group. The electron-rich
substituents, as the pyrrole in 15h, donated electrons to the adenine ring increasing the
basicity of N-1 and, therefore, its pKa. A red shift (bathochromic effect) in λem was
observed for 15e and 15g as the pH increased. This effect could be due to a stronger
vibrational relaxation occurring at basic pH, because of the lack of H-bonds occurring at
acidic pH (Table 3.7, Figure 3.17). In contrast, λem for the other substituted adenine
rings did not undergo any substantial shift with the pH (Table 3.7).
λem/nm
cmpd R (substituent on C-8) pH 3 pH 4 pH 6 pH 7 pH 8 pH 9
15b Phenyl 385 382 383 383 383 383
15c 3-(N-Boc-aminomethyl)phenyl 383 383 383 384 384 384
15e Pyridin-3-yl 374 390 404 404 406 406
15g 2,4-DMT-pyrimidin-5-yl 380 387 413 413 413 415
15h Pyrrol-2-yl 375 375 375 375 375 375
Table 3.7 λem maxima of 15b,c,e,g,h at different pH. Conditions: NaOAc/AcOH buffer (50 mM) was used
for pH 3 and 4; KH2PO4/NaOH buffer (50 mM) was used for pH 6; Tris/HCl buffer (50 mM) was used
for pH 7, 8 and 9. Fluorescence emission spectra at λex maxima (see Table 3.5), rt.
Chapter 3
106
Figure 3.17 Network of H-bonds around 15e and 15g, responsible for the red shift of λem in 15e and 15g
with the increase of the pH, and the hypochromic effect at basic pH in 15g. Conditions: NaOAc/AcOH
buffer (50 mM) was used for pH 3 and 4; KH2PO4/NaOH buffer (50 mM) was used for pH 6; Tris/HCl
buffer (50 mM) was used for pH 7, 8 and 9. The compound concentrations were respectively: 3.6 μM for
15e and 3.5 μM for 15g. Fluorescence emission spectra were recorded at λex maxima (see Table 3.5), rt.
Emission spectra areas were normalized to the highest value for each compound.
3.7 Fluorescence quenching in 8-aryl/heteroaryl NAD+ derivatives
3.7.1 Fluorescence basis for internal quenching in NAD+ derivatives
As mentioned in 3.6, clues for a dynamic/static fluorescence quenching due to the
intramolecular interactions between the nicotinamide and the 8-aryl/heteroaryl adenine
rings of the corresponding NAD+ derivatives were found comparing values in water
and methanol. Indeed, the values suggested a fluorescence quenching occurring in
water, where the β-NAD+ scaffold would assume a folded conformation with the
hydrophobic aromatic rings in close proximity with each other. Methanol would disrupt
these interactions solvating the aromatic rings and allowing the passage of β-NAD+
from a folded to an extended conformation with re-establishment of the fluorescence.64,
209
A thorough investigation was undertaken to explore this phenomenon in the series of 8-
aryl/heteroaryl NAD+ derivatives synthesized in this project. Similar studies have been
previously undertaken with another important fluorescent NAD+ derivative, ε-NAD
+,87
already introduced in Chapter 1 (Figure 3.18). This compound and its derivatives (ε-
AMP, ε-ADP, ε-ATP, ε-ADPR, cεADPR) have been extensively studied since the early
1970s, and have found wide applications as fluorescent probes for several enzymes, due
to their characteristic fluorescent features.83, 210
Indeed, etheno-derivatives are
0
0.2
0.4
0.6
0.8
1
1.2
300 350 400 450 500
No
rma
lize
d F
luo
resc
en
ce E
mis
sio
n
wavelength [nm]
pH3
pH4
pH6
pH7
pH8
pH9
0
0.2
0.4
0.6
0.8
1
1.2
300 350 400 450 500
No
rma
lize
d F
luo
resc
en
ce E
mis
sio
n
wavelength [nm]
pH3
pH4
pH6
pH7
pH8
pH9
λemλem
Chapter 3
107
characterized by λex and λem not interfering with those of biological systems (Table 3.8).
Moreover, their high values make them very good fluorescent probes.
cmpd λex/nm λem/nm Stokes shifta/cm
-1
ε-NAD+ 83
306 410 8289 0.028
ε-AMP 306 400 7680 0.56
Tryptophan181
295 353 5570 0.13
Tyrosine181
275 304 3469 0.14
Phenylalanine181
260 282 3000 0.02 a Stokes shift = (1/ex - 1/em).
Table 3.8 Spectroscopic properties of ε-NAD+ and fluorescent amino acids.
In the case of ε-NAD+, a low quantum yield value is due to a strong internal
fluorescence quenching occurring between the stacked ε-adenine and nicotinamide rings
in the ε-NAD+ structure. These stacking interactions can be disrupted by chemical or
enzymatic removal of the nicotinamide ring with consequent fluorescence increase.
Indeed, the quantum yield value of ε-AMP is ~ 20 fold greater than the corresponding ε-
NAD+ (Table 3.8, Figure 3.18).
83, 87, 88 The internal quenching occurring in the intact ε-
NAD+ compared to the ε-adenine moiety is the basis for the development of
fluorescence-based assays for NAD+-consuming enzymes,
7 that use β-NAD
+ as a
substrate, cleaving the high-energy bonds present in its structure (i.e. the pyrophosphate
and N-glycosidic bond). A short review of the main assays for NAD+-consuming
enzymes based on the fluorescence changes of ε-NAD+ has been presented in 1.4.2.
The fluorescence quenching in ε-NAD+ is due to a mixed dynamic/static quenching.
The dynamic quenching is mainly due to intramolecular collisional interactions of the
excited molecule and, therefore, it affects the quantum yield and the fluorescence
lifetime. In contrast, the static quenching is due to the formation of dark, not emissive
complexes between the stacked bases in the ground state and, therefore, it affects the
quantum yield but not the fluorescence lifetime. It is possible by measurement of
fluorescence lifetimes to distinguish the two type of quenching and their contributions;
it has been shown that ε-NAD+ is subjected to both, with prevalence of the static
quenching. Moreover, it has been calculated that ε-NAD+ at 25 °C in water is present in
a 45 ± 5% stacked conformation.88
Chapter 3
108
Figure 3.18 ε-NAD+ and its fluorescence internal quenching.
Different studies were undertaken to prove the existence of an intramolecular
fluorescence quenching in the series of fluorescent 8-aryl/heteroaryl NAD+ derivatives.
Interesting information could be deducted from the comparison of quantum yield values
between AMP and NAD+ derivatives (Table 3.9). The quantum yield values of 8-
aryl/heteroaryl AMP derivatives were higher than the corresponding NAD+ derivatives,
with the only exception of 8-(pyridin-3-yl) AMP (9e) and its corresponding NAD+
derivative (15e).172
As for ε-NAD+, these differences in quantum yield could occur
because of an intramolecular fluorescence quenching due to stacking and collisional
interactions between the fluorophore and the nicotinamide ring. The larger the
differences, the stronger would be the interactions between the two aromatic rings. 8-
(Pyrrol-2-yl) AMP (9h) and its corresponding NAD+ derivative (15h) showed the
largest difference in water amongst the NAD+ derivatives synthesized. This effect
could be due to a stronger planarity of the 8-(pyrrol-2-yl) adenine ring, which facilitated
the stacking interactions with the nicotinamide ring. When values for this couple
were measured in methanol, difference decreased since the organic solvent helped the
solvation and the separation of the hydrophobic aromatic rings in 15h with re-
establisment of the fluorescence. In contrast, 8-(pyridin-3-yl) AMP (9e) and its
corresponding NAD+ derivative (15e) did not show any difference either in water or
methanol, indicating a possible lack of planarity in the 8-(pyridin-3-yl) adenine ring
and, consequently, poor stacking interactions with the nicotinamide ring.
Weak fluorescence Strong fluorescence!
Chapter 3
109
cmpd scaffold Position of
substitution R (substituent) solvent AMP NAD AMP/NAD
NAD+83
NAD n.a. n.a. Water 0.56 0.028 20
AMP83
AMP
14b
8b
NAD 6 Pyrrol-2-yl Water 0.0041 0.0012 3.42
AMP
15b NAD 8 Phenyl
Water 0.669 0.2067 3.24
9b AMP MeOH 0.225 0.080 2.81
15c NAD 8
3-(N-Boc-
aminomethyl)phenyl
Water 0.888 0.150 5.92
9c AMP MeOH 0.639 0.148 4.32
15e NAD 8 Pyridin-3-yl
Water 0.0034 0.0035 0.97
9e AMP MeOH 0.021 0.016 1.31
15g NAD 8
2,4-DMT-pyrimidin-5-
yl
Water 0.021 0.0026 8.08
9g AMP MeOH 0.147 0.0054 27.22
15h NAD 8 Pyrrol-2-yl
Water 0.231 0.0053 43.58
9h AMP MeOH 0.272 0.055 4.94
n.a. not applicable.
Table 3.9 Comparison of values in water and methanol for ε-AMP/ε-NAD+, and aryl/heteroaryl
adenine-modified AMP/NAD+ derivatives synthesized in this study.
To further prove the existence of the internal quenching in 8-aryl/heteroaryl NAD+
derivatives and its decrease passing from water to an organic solvent, a methanol
titration of aqueous solutions of 8-aryl/heteroaryl NAD+ derivatives was performed
(Figure 3.19). In fact, even if the trends did not completely agree with values
calculated in methanol for the different compounds, a general fluorescence increase was
observed with the increase of methanol percentage, as expected from the disruption of
the internal quenching in organic solvent.211
In addition, a blue shift in λem maxima for
the different NAD+ derivatives was observed with the increase of the methanol
percentage because of a lower solvent relaxation occurring in methanol compared to
water (Table 3.10). However, the shift was so small that it did not substantially modify
the Stokes shift values, as observed in 3.6, making the NAD+ derivatives quite
insensitive to slight environmental changes.
Chapter 3
110
Figure 3.19 MeOH titration of 15c,e,g,h. Conditions: aqueous solutions of the different NAD+ derivatives
with several MeOH percentage (ranging from 10 to 90% MeOH with an interval of 20%) were prepared
and their fluorescence emission spectra recorded at λex maxima (see Table 3.5), rt. The fluorescence
emission values at λem maxima for the spectra at different MeOH percentage were normalized for those at
λem maxima in H2O for each compound; then, the normalized values were plotted as function of the
solvent composition.
λem/nm at different MeOH/H2O (%)
cmpd R (substituent on C-8) 0 10 30 50 70 90
15c 3-(N-Boc-aminomethyl)phenyl 384 382 382 382 379 380
15e Pyridin-3-yl 406 415 405 401 399 407
15g 2,4-DMT-pyrimidin-5-yl 411 414 410 410 409 409
15h Pyrrol-2-yl 375 374 373 373 372 370
Table 3.10 λem maxima of 15b,d,f,g at different MeOH percentage. Conditions: aqueous solutions of the
different NAD+ derivatives with several MeOH percentage (ranging from 10 to 90% MeOH with an
interval of 20%) were prepared and their fluorescence emission spectra recorded at λex maxima (see Table
3.5), rt.
Finally, a chemical hydrolysis of the fluorescent 8-aryl/heteroaryl NAD+ derivatives
was performed to give conclusive evidence of the internal fluorescence quenching due
to the intramolecular interactions between the fluorescent substituted adenine and
nicotinamide rings. As discussed in 1.1, the nicotinamide N-glycosidic bond can be
hydrolysed in 0.1 – 1 N NaOH with a first order reaction in hydroxide.8, 9, 11
The
removal of NaM in 8-aryl/heteroaryl NAD+ derivatives would result in the formation of
the corresponding 8-aryl/heteroaryl ADPR derivatives, with re-establishment of the
fluorescence of the substituted adenine rings.
0
1
2
3
4
5
6
7
8
0 10 20 30 40 50 60 70 80 90 100
No
rma
lize
d F
luo
res
ce
nc
e E
mis
sio
n
% MeOH in H2O
3-(Boc-aminomethyl)phenyl (15c)
Pyridin-3-yl (15e)
2,4-DMT-pyrimidin-5-yl (15g)
Pyrrol-2-yl (15h)
Chapter 3
111
The chemical hydrolysis for all 8-aryl/heteroaryl NAD+ derivatives was followed by
fluorescence (Figure 3.20) and, in the case of 8-(pyrrol-2-yl) NAD+ (15h), also by
HPLC (Figure A1).172
As expected, a fluorescence increase was observed during the
hydrolysis of all NAD+ derivatives examined. However, the fluorescence increase was
higher for 15h, which showed the strongest fluorescence internal quenching amongst
the NAD+ derivatives synthesized. The HPLC time-course for the basic hydrolysis of
15h confirmed the formation of NaM and the corresponding ADPR derivative, 23.
Figure 3.20 Fluorescence time-course for the basic hydrolysis of NAD+ derivatives 15b,c,e,g,h.
Conditions: 0.8 μM in 0.1 N NaOH, rt; for λex and λem maxima, see Table 3.5.
3.7.2 NMR basis for internal quenching in NAD+ derivatives
To investigate if dynamic/static quenching was occurring in aryl/heteroaryl adenine-
modified NAD+ derivatives, NMR studies were undertaken. Indeed, the occurrence of
π-stacking interactions between aromatic rings can influence the chemical shifts of their
protons because of a “shielding-increasing” ring current. Therefore, an upfield shift
should be observed with the increase of π-stacking interactions, which shield the
aromatic protons; and viceversa, a downfield shift should occur when the π-stacking
interactions decrease and the protons are deshielded.
0
1
2
3
4
5
6
0 200 400 600 800 1000 1200 1400 1600 1800
No
rma
lize
d F
luo
res
ce
nce
Em
iss
ion
time [sec]
Phenyl (15b)
3-(Boc-aminomethyl)phenyl (15c)
Pyridin-3-yl (15e)
2,4-DMT-pyrimidin-5-yl (15g)
Pyrrol-2-yl (15h)
Chapter 3
112
This phenomenon has been largely studied on nucleosides and nucleotides which self-
associate by intermolecular π-stacking interactions depending on their concentration in
aqueous solution.212, 213
The more diluted the solution, the less self-association is
observed between the nucleobases. In the case of β-NAD+ and NAD
+ derivatives, the
prevalent intramolecular π-stacking interactions are concentration-independent; only at
high concentrations, also intermolecular associations start to occur.214-218
Consequently,
to show the occurrence of intramolecular π-stacking interactions in diluted aqueous
solutions, a titration experiment monitored by 1H NMR was set up to observe the effect
of increasing percentages of acetonitrile on the protons of the natural β-NAD+ and 8-
(pyrrol-2-yl) NAD+ (15h). Changes in the chemical shifts of the protons of the two
molecules should show the passage of the β-NAD+ scaffold from a folded to an open
conformation. This experiment was the analogue of the methanol titration performed on
8-aryl/heteroaryl NAD+ derivatives and monitored by fluorescence in Figure 3.19. As
expected, a downfield shift in the aromatic protons occurred with the increase of
acetonitrile in both the natural β-NAD+ and 15h, with a stronger effect in the case of
15h. Indeed, the 8-(pyrrol-2-yl) adenine ring could be able to establish stronger π-
stacking interactions than the adenine itself with the nicotinamide ring because of a
more extended ring current and strong planarity.
Following the titration of β-NAD+, 2N-H and 2-H showed a significant downfield shift
with the increase of acetonitrile in solution, compared to a smaller shift for 4N-H and
5N-H. In contrast, 6N-H and 8-H showed an upfield shift (Figure 3.21). Strong
interactions between the pyrimidine part of the adenine ring and the nicotinamide would
occur in water more than in acetonitrile, facilitated by the formation of H-bonds; the
result would be the substantial shielding of 2N-H and 2-H. Conversely, 6N-H and 8-H,
because of their positions on the aromatic rings, would be only partially involved in the
shielding interactions occurring between the six-membered rings; this would result in a
deshielding effect stronger in water than in acetonitrile.
Chapter 3
113
Figure 3.21 1H NMR analysis of aromatic protons in β-NAD
+ with increasing percentage of ACN.
indicates the difference in chemical shift between the highest and the lowest ACN percentage: values are
given in ppb (part per billion: 1000 ppb = 1 ppm). Conditions: solutions in D2O of β-NAD+ (0.0025 M)
with several ACN percentage (ranging from 10 to 60% anhydrous ACN with an interval of 10%) were
prepared and their 1H NMR spectra recorded at rt. The spectra were referenced to the protons of CH3 in
ACN (H 2.06).
Since the solvation of the aromatic rings in organic solvents promotes the passage of β-
NAD+ from a folded to an open conformation, a change in the conformation of the
riboses and, consequently, in the chemical shifts and J coupling constants of their
protons (Figure 3.22, Figure 3.23) should be observed.
The chemical shifts and J coupling constants of 1′-H and 1′′-H were analysed. 1′-H and
its J1ꞌ,2ꞌ did not undergo any substantial change with the solvent composition. On the
other hand, 1′′-H did undergo an upfield shift, or a shielding effect, with increasing
percentage of acetonitrile. This effect could be due to a change in the conformation of
the nicotinamide ribose, which would bring 1′′-H far apart from the positively charged
nicotinamide ring. In addition, J1ꞌꞌ,2ꞌꞌ decreased with increasing acetonitrile percentage,
indicating the preference of the nicotinamide ribose toward a N-conformation in the β-
NAD+ extended conformation (see 3.1).
Aromatic
proton (ppb)
2N-H 19.3
6N-H -9.8
4N-H 4.3
2-H 20.2
8-H -22.6
5N-H 11.4
8.615
8.62
8.625
8.63
8.635
8.64
9.4
9.41
9.42
9.43
9.44
9.45
9.46
9.47
0 10 20 30 40 50 60 70
(p
pm
) 2
-H
(
pp
m)
2N-H
% ACN in water
2N-H
2-H
Chapter 3
114
Figure 3.22 1H NMR analysis of 1′-H and 1′′-H in β-NAD
+ with increasing percentage of ACN.
indicates the difference in chemical shift between the highest and the lowest ACN percentage: values are
given in ppb (part per billion: 1000 ppb = 1 ppm). Conditions: solutions in D2O of β-NAD+ (0.0025 M)
with several ACN percentage (ranging from 10 to 60% anhydrous ACN with an interval of 10%) were
prepared and their 1H NMR spectra recorded at rt. The spectra were referenced to the protons of CH3 in
ACN (H 2.06).
In Figure 3.23, the NMR peaks for 1′-H and 1′′-H are shown at different acetonitrile
percentage. 1′′-H showed a substantial upfield shift compared to 1′-H with increasing
acetonitrile percentage. The 1′′-H peak, during its upfield shift, overlapped the 1′-H
peak at 20% of acetonitrile.
Figure 3.23 1H NMR peaks of 1′-H and 1′′-H in β-NAD
+ with increasing percentages of ACN.
Conditions: solutions in D2O of β-NAD+ (0.0025 M) with several ACN percentage (ranging from 10 to
60% anhydrous ACN with an interval of 10%) were prepared and their 1H NMR spectra recorded at rt.
The spectra were referenced to the protons of CH3 in ACN (H 2.06).
Sugar protons (ppb)
1 -H 0.9
1 -H -70.2
J coupling J (Hz)
J1ꞌ,2ꞌ 0.16
J1ꞌꞌ,2ꞌꞌ -0.44
6.09
6.1
6.11
6.12
6.13
6.14
6.15
6.16
6.17
0 10 20 30 40 50 60 70
(p
pm
)
% ACN in water
1''-H
1'-H
NAD in 10% ACN
NAD in 20% ACN
NAD in 40% ACN
NAD in 50% ACN
NAD in 60% ACN 1′′1′
1′1′′
Chapter 3
115
In 15h, a downfield shift with increasing acetonitrile percentage occurred for all
aromatic signals with stronger or weaker intensity, indicating an extended “shielding-
increasing” ring current involving all aromatic protons (Figure 3.24). However, 2-H,
which in β-NAD+ was undergoing a strong downfield shift, in 15h showed only a very
weak shift. This effect could be due to the deshielding effect played by the close
phosphate group, since the adenine ring assumed in 15h a syn conformation (Figure
3.25).
Figure 3.24 1H NMR analysis of aromatic protons in 15h with increasing percentage of ACN.
indicates the difference in chemical shift between the highest and the lowest ACN percentage: values are
given in ppb (part per billion: 1000 ppb = 1 ppm). Conditions: solutions in D2O of 15h (0.0025 M) with
several ACN percentage (ranging from 10 to 60% anhydrous ACN with an interval of 10%) were
prepared and their 1H NMR spectra recorded at rt. The spectra were referenced to the protons of CH3 in
ACN (H 2.06).
In 15h, 1′-H and 1′′-H, conversely to their behaviour in β-NAD+, showed a downfield
shift, or a deshielding effect, with increasing acetonitrile percentage (Figure 3.26). This
effect could indicate the involvement of these protons in the “shielding-increasing” ring
current, stronger in 15h than in β-NAD+. In addition, J1ꞌ,2ꞌ and J1ꞌꞌ,2ꞌꞌ decreased with
increasing acetonitrile percentage. Therefore, during the passage of 15h from a folded to
an open conformation, both riboses puckered toward a N-conformation (see 3.1),
especially the nicotinamide ribose.
Aromatic
proton (ppb)
2N-H 119.9
6N-H 81
4N-H 87.7
2-H 28.1
5N-H 63.3
5py-H 11.8
3py-H 60.9
4py-H 9.9
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
9
8.5
8.6
8.7
8.8
8.9
9
9.1
9.2
9.3
9.4
0 10 20 30 40 50 60 70 80 90 100
(
pp
m)
2-H
, 5
N-H
(p
pm
) 2
N-H
, 6N-H
, 4N-H
% ACN in water
2N-H
6N-H
4N-H
2-H
5N-H
Chapter 3
116
Figure 3.25 β-NAD+ and 15h folded conformations.
Figure 3.26 1H NMR analysis of 1′-H and 1′′-H in 15h with increasing percentage of ACN. indicates
the difference in chemical shift between the highest and the lowest ACN percentage: values are given in
ppb (part per billion: 1000 ppb = 1 ppm). Conditions: solutions in D2O of 15h (0.0025 M) with several
ACN percentage (ranging from 10 to 60% anhydrous ACN with an interval of 10%) were prepared and
their 1H NMR spectra recorded at rt. The spectra were referenced to the protons of CH3 in ACN (H 2.06).
From the 2D NOESY experiment on 15h, no particular NOE signals were observed,
with the exception of that correlating 1′-H and 3pyrrole -H, as expected because of the
syn conformation of the adenine and the consequent proximity of the two protons
(Figure 3.25). Unfortunately, the molecular weight of 15h is in the range where the
NOE signal should be theoretically zero; consequently, the 2D NOESY experiment did
not allow observation of any interaction between the aromatic bases.
Sugar protons (ppb)
1 -H 115.1
1 -H 58.9
J coupling J (Hz)
J1ꞌ,2ꞌ -80
J1ꞌꞌ,2ꞌꞌ -200
5.8
5.85
5.9
5.95
6
6.05
6.1
0 20 40 60 80
(p
pm
)
% ACN in water
1''-H
1'-H
Chapter 3
117
Of course, it is difficult to predict only from the 1H NMR experiments the presence of
β-NAD+ and its derivatives in solution in a folded or open conformation. Many studies
have been undertaken to find a univocal solution using different techniques; however,
contrasting opinions have been often reached.64, 214-216
Surely, intramolecular
interactions between the aromatic bases are present in aqueous solutions of β-NAD+,
due to stacking or just proximity effects. However, it is not possible to exclude the
presence of intermolecular stacking interactions depending on the concentration. In
addition, the interactions between the nucleobases could occur in many different ways,
depending on the orientation of the aromatic faces.214
The results shown here seem at
least to confirm the presence of interactions between the aromatic bases in aqueous
solutions and, in particular, a stronger effect occurring in 15h compared to β-NAD+.
This is in agreement with what we would expect from a more extended π-electron
delocalization occurring in the 8-(pyrrol-2-yl) adenine, which increases the eventual π-
stacking interactions with the nicotinamide ring.
3.8 Conclusions
In contrast to the non-emissive adenine derivatives modified in position 2 or 6 with
aryl/heteroaryl substituents, the fluorescence features of 8-aryl/heteroaryl adenine
derivatives make them interesting compounds from the perspective of the applicability
as biochemical tools to investigate enzymatic structures, bindings and activities. Indeed,
since the development of the ε-adenine derivatives and their versatile biological
applications, not many other isomorphic fluorescent adenine derivatives have been
developed with the same purpose of fluorescent probes. However, in this series of
isomorphic emissive 8-aryl/heteroaryl adenine derivatives a new fluorophore class can
be identified with tunable fluorescent properties and, in some cases, as will be shown
later, unique behaviours toward certain types of enzymes. A comparison between ε-
adenine and 8-aryl/heteroaryl adenine derivatives is shown in Table 3.11 and Figure
3.27.
Chapter 3
118
ε-adenine derivatives 8-aryl/heteroaryl adenine derivatives
Fixed modification (etheno bridge) Variable modification (aryl/heteroaryl substituent)
N-1 and NH2 blocked N-1 and NH2 free for H-bonds
Adenine in anti conformation Adenine in syn conformation
λex = 306 nm; λem = 410 nm (ε-NAD+) λex = 309 nm; λem = 374 nm (15h)
AMP/NAD = 20 (ε-NAD+) AMP/NAD = 44 (15h)
Table 3.11 Comparison between ε-adenine and 8-aryl/heteroaryl adenine derivatives.
A characteristic feature of 8-aryl/heteroaryl adenine derivatives is their highly variable
structure, due to the large range of substituents that can be introduced on C-8 of the
adenine ring by Suzuki-Miyaura cross-coupling. As shown later in the chapter, by
changing the substituent, it is possible to modulate the fluorescent properties of the
compounds. Therefore, it would be theoretically possible to predict the fluorescent
properties depending on the nature of the introduced substituents. This structural
variability is not applicable to the ε-adenine, where the etheno bridge, extending the
aromaticity of the modified adenine ring, is responsible for the fluorescent properties of
the compounds. No substantial modifications can be performed on its structure to
change or tune the fluorescent features of this fluorophore class, resulting therefore in a
big disadvantage compared to emissive 8-aryl/heteroaryl adenine derivatives.
In addition to the fact that ε-adenine structure cannot be modified to increase
fluorescence, the introduction of the etheno bridge between the exocyclic amino group
and N-1 on the pyrimidine ring blocks two important moieties responsible for enzyme
recognition and binding. Conversely, the new fluorophore class leaves the adenine
structure unaltered making the compounds more similar to the natural substrate and,
therefore, more likely to interact with the enzymes.
The two fluorophore classes differ also in the orientation of the modified adenine rings
related to their riboses; indeed, ε-adenine prefers an anti conformation, while 8-
aryl/heteroaryl adenine a syn conformation. This difference could have separate
influences on their biological activities toward NAD+-consuming enzymes, as will be
discussed in Chapter 4 and 5.
Chapter 3
119
Figure 3.27 Structures for -adenine and 8-aryl/heteroaryl adenine derivatives.
A common feature between the two fluorophore classes is represented by the internal
fluorescence quenching occurring in the NAD+ derivatives. Based on the example of ε-
NAD+ and depending on its ability to act as a NNS for NAD
+-consuming enzymes, 8-
(pyrrol-2-yl) NAD+ (15h), the most promising fluorescent 8-aryl/heteroaryl NAD
+
derivative synthesized in Chapter 2, will be used to set up fluorescence-based assays for
NAD+-consuming enzymes. Indeed, this compound, as seen with ε-NAD
+, is
characterized by λex and λem not interfering with the protein fluorescence; in addition, its
fluorescence quenching, ~ 2 fold stronger than ε-NAD+, could allow an increase in
sensitivity of the fluorescence-based assays.
Chapter 4
120
4 EVALUATION OF 8-ARYL/HETEROARYL NAD+ DERIVATIVES AS FLUORESCENT PROBES FOR NAD+-CONSUMING ENZYMES
As mentioned in Chapter 1, NAD+-consuming enzymes are fundamental in all
organisms as they are involved in many major biological processes. These enzymes use
β-NAD+ as a substrate/co-substrate, cleaving the high energy bonds present in its
structure; then, the energy released can be used for the synthesis of signalling molecules
or transferred to other acceptors to catalyze further reactions. Several different types of
NAD+-consuming enzymes are present in cells and have a variety of functions, often
related to each other. The knowledge of their activities is therefore crucial to understand
many biological processes and their pathological implications.
This chapter will review activities and reaction mechanisms of some representatives of
major classes of NAD+-consuming enzymes, in particular those belonging to the ecto-
enzymes class, such as NPPs, ADPRCs and NADases, responsible for the extracellular
β-NAD+ homeostasis. Since β-NAD
+ is the only substrate involved in their catalytic
reactions, they will be used as a model for the proof of concept of the use of NAD+
derivatives as biochemical tools.
In particular, the biological evaluation of 15h as a fluorescent NNS for the selected
NAD+-consuming enzymes will be provided, and its use for the development of
fluorescence-based assays for their monitoring will be assessed. Indeed, this compound,
amongst all NAD+ derivatives synthesized in Chapter 2, has shown the best fluorescent
features to compete with ε-NAD+ as an alternative biochemical probe for the monitoring
of the NAD+-consuming enzyme activities. Its fluorescence stability between pH 6 and
9, the pH working range for several NAD+-consuming enzymes, and its λex and λem not
overlapping with those of biological systems, make it a good fluorophore for biological
applications. In addition, the internal fluorescence quenching occurring in 15h is ~ 2
fold bigger than that occurring in ε-NAD+, possibly making it a more sensitive probe in
fluorescence-based assays. Moreover, the intact adenine structure in 15h could be an
advantage compared to ε-NAD+ in terms of interactions and recognition by NAD
+-
consuming enzymes.
Chapter 4
121
4.1 Nucleotide pyrophosphatases (NPPs)
Nucleotide pyrophosphatases/phosphodiesterases (NPPs) belong to the same class of
enzymes, able to catalyze the hydrolysis of pyrophosphate or phosphodiester bonds of a
variety of nucleotides to give nucleoside 5′-monophosphates.16
A broad range of
possible substrates are accepted by NPPs.
Mammalian NPPs are also called ecto-NPPs (ENPPs) since most of them are ecto-
enzymes, or enzymes situated on the outer membrane of the cell and characterized by
having their catalytic site in the extracellular space. The activity of these enzymes in the
extracellular space seems to be fundamental to regulate the extracellular balance
between nucleotides and nucleosides connected to a variety of events, such as
nucleotide/nucleoside signalling (purinergic modulation signalling), the salvage
pathway of many nucleotides, bone and cartilage mineralization, and insulin
signalling.16, 17, 219, 220
Because of these activities, an increased interest toward
mammalian ENPPs has developed to fully characterize and define their structure, active
site and reaction mechanism.
ENPPs share the same reaction mechanism already identified for snake venom
phosphodiesterases:48
it is a two-step mechanism, involving a covalent enzyme-
nucleotidylate intermediate by a threonine catalytic residue (Scheme 4.1).221
The ENPPs
catalytic site contains two divalent cations, coordinating the phosphate group of
pyrophosphate or phosphodiester bonds of the incoming nucleotides, BppB′ and BpB′
respectively. A nucleophilic residue, such as threonine (T256) in human NPP1, attacks
the phosphate with consequent release of a leaving group, such as a nucleotide B′p or a
nucleoside B′, and formation of a covalent enzyme-nucleotidylate intermediate (I).
Therefore, a molecule of water, activated by coordination to the cation, would attack the
phosphate, regenerating the catalytic threonine and releasing the product, a nucleoside
5′-monophosphate, Bp.16, 222
Chapter 4
122
Scheme 4.1 Catalytic mechanism of ENPPs.
The observed dependence of the efficiency of the pyrophosphatase reactions on divalent
cations in the reaction buffer can be explained by the necessary presence in the active
site of two cations to catalyze the reaction.223, 224
In addition, the optimum basic pH for
the pyrophosphatase reaction helps to speed the second reaction step, where a more
nucleophilic hydroxide, instead of a neutral molecule of water, would attack the
phosphate group.
In contrast to the several studies performed on snake venom phosphodiesterases and
ENPPs, little information is known about the commercially available nucleotide
pyrophosphatase from Crotalus adamanteus venom. This enzyme has wide application
in the chemo-enzymatic synthesis of nucleotide derivatives with quite good yields225
but
has not been fully characterized. Here, its activity as a nucleotide pyrophosphatase, its
properties and reaction mechanism will be discussed in accordance with its assumed
similarity to other well described nucleotide pyrophosphatases/phosphodiesterases.
Chapter 4
123
4.1.1 Fluorescence-based assay for nucleotide pyrophosphatase
Compound 15h was tested as a NNS of the commercially available nucleotide
pyrophosphatase (NPP) from Crotalus adamanteus venom. This enzyme catalyzes the
cleavage of β-NAD+ on the pyrophosphate bond to produce AMP and β-NMN.
The enzymatic reaction was monitored by HPLC to check the suitability of 15h as a
substrate for NPP; in fact, the enzyme recognised the NNS and catalyzed its hydrolysis
on the pyrophosphate bond generating β-NMN and 9h,172
as confirmed by the use of
standards for the two reaction products (Figure 4.1).
Figure 4.1 HPLC time-course of the enzymatic reaction of NPP from Crotalus adamanteus venom and
15h. Conditions: NPP (0.01 U/mL), 15h (24.74 μM), MgCl2 (10 mM), Tris/HCl buffer (50 mM, pH 8), 30
ºC; all concentrations are final concentrations. Data from single experiment, peak area vs time [min], data
plotting to a 1st order rate equation with Grafit5.
Since NPP was able to recognise and convert 15h into 9h, the opportunity for
monitoring the enzymatic reaction by fluorescence was investigated. Owing to the
fluorescence internal quenching occurring in 15h compared to 9h (Chapter 3, see 3.7.1),
a fluorescence increase due to the production of 9h from the cleavage of 15h was
observed with the progress of the enzymatic reaction (Figure 4.2), proving the
possibility to prepare fluorescence-based assays to monitor NPP activity, similar to
those developed with ε-NAD+.87, 89
Chapter 4
124
Kinetic experiments for the estimation of kinetic parameters, as Km and Vmax, for NPP
toward the new substrate 15h were set up. After confirming the linearity of the
fluorescence response across a range of concentrations for both 15h and its enzymatic
product 9h, initial reaction velocities were measured as a function of NPP concentration
at a fixed 15h concentration to explore the rate changes with NPP concentration
(Appendix 8, Figure A2). Initial velocities were found to be proportional at low NPP
concentration, before the substrate 15h became the limiting factor. A suitable
enzyme/substrate ratio was established to have smooth enzymatic reaction rates from
which to calculate linear initial velocities. Afterwards, the substrate concentration was
varied to generate a saturation curve for the determination of the value of Km.
At this stage, the experiment set-up required a considerable optimization to establish the
appropriate conditions to reduce and control the variability affecting the results and
attributed to several factors influencing the fluorescence signal. Major issues identified
were the temperature changes within the wells of the microplate, due to the addition of
the cold enzyme solution at the beginning of the reaction, and the instability of the
fluorescence signal. To address these problems, a few precautions were taken, such as
the addition of all components of the reaction into the microplate at room temperature
and always in the same order; a fluorescence stabilization period was allowed before
starting the reaction; shaking of the microplate occurred before each cycle of
measurement; use of 120 μL total volume in the wells (100 – 250 μL is the volume
range suggested for the NUNC 96 plate). The introduction of positive and negative
controls into the microplate was useful to check the variability sources in the
experiments. The buffer alone was used as a negative control, while the reaction
product, 9h, at a fixed concentration was used as a positive control. Their variability
was checked across the plate after the fluorescence signal stabilization and during the
reaction time to guarantee a uniform fluorescence signal. No particular fluorescence
trends were noticed across the plate, indicating the suitability of NUNC 96 plate for
these experiments.
Chapter 4
125
All experiments were run at least in duplicate. The raw data showed time-courses of
fluorescence intensity, expressed in RFU [relative fluorescence unit], versus time [sec].
The fluorescence difference between the positive control (9h at a fixed concentration)
and the negative control (buffer) allowed conversion of the fluorescence intensity into
9h concentration [μM], necessary to estimate Vmax (Figure 4.2).
Figure 4.2 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h. a) fluorescence time-
course as fluorescence emission [RFU] vs time [sec]; b) fluorescence time-course as 9h concentration
[μM] vs time [sec]. Conditions: NPP (0.007 U/mL), 15h (0.2 – 6 μM), MgCl2 (10 mM), Tris/HCl buffer
(50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%. Enzyme addition at 420 sec. Positive
control: 9h (3 μM). Negative control: buffer; all the concentrations are final concentrations. Each data
point is the mean of experiments run in duplicate, data plotting with Excel.
Variation of initial velocity with 15h concentration was well described by the
Michaelis-Menten equation (Figure 4.3), and Lineweaver-Burk and Hanes-Woolf plots
(Appendix 8, Figure A3).172
Values for Km and Vmax of NPP related to 15h were
determined by non-linear regression analysis of the Michaelis-Menten plot, resulting in
an average Km value of 1.5 ± 0.4 μM and Vmax of 0.001 ± 0.0001 μmol·min-1
·mg-1
.
0
0.5
1
1.5
2
2.5
3
0 1000 2000 3000 4000 5000
[9h
] m
M
time [sec]
[15h] 0.2uM
[15h] 1uM
[15h] 2uM
[15h] 3uM
[15h] 4uM
[15h] 5uM
[15h] 6uM
-1000
1000
3000
5000
7000
9000
11000
0 1000 2000 3000 4000 5000
Flu
ore
sc
en
ce E
mis
sio
n [
RF
U]
time [sec]
[15h] 0.2uM
[15h] 1uM
[15h] 2uM
[15h] 3uM
[15h] 4uM
[15h] 5uM
[15h] 6uM
neg contr
pos contr
a b
NPP
Chapter 4
126
The presence of Mg2+
in the buffer was required to stimulate the enzymatic reaction;223,
224 indeed, Vmax showed a drop of ~ 3.5 fold when the reaction was carried out in
absence of MgCl2 (data not shown).
Figure 4.3 Michaelis-Menten plot of NPP from Crotalus adamanteus venom with 15h. Conditions: NPP
(0.007 U/mL), 15h (0.2 – 5.2 μM), MgCl2 (10 mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm,
λem = 410 nm, gain 15%; all concentrations are final concentrations. Initial velocity values (V0) were
calculated from the corresponding time-courses measuring the fluorescence signals generated when 10%
product was formed. Each data point is the mean (± SD) of experiments run in duplicate, data plotting
with Excel.
To assess the NPP site of 15h binding and transformation, experiments were performed
to evaluate the impact of β-NAD+, AMP and β-NMN on the rate of hydrolysis for 15h
by NPP. Competition experiments with β-NAD+ showed a decrease of NPP activity
related to 15h for increasing quantity of β-NAD+ in solution: at concentrations of β-
NAD+ above ca 250 μM, there was no significant indication of catalysis. An IC50 value
of 12.8 ± 3.8 μM was estimated for β-NAD+ (Figure 4.4).
For greater insight into the mode of inhibition exhibited by β-NAD+ toward the activity
of NPP related to 15h, kinetic analysis was performed for a range of β-NAD+
concentrations (Figure 4.5, Appendix 8, Figure A4). From the data analysis, β-NAD+
turned to be a competitive inhibitor of 15h; indeed, Vmax did not change with increasing
β-NAD+ concentrations, while the Km value of 15h related to NPP increased because of
a lower affinity for the binding site.
0
0.0001
0.0002
0.0003
0.0004
0.0005
0.0006
0.0007
0.0008
0.0009
0.001
0 1 2 3 4 5 6
V0
[mM
/sec]
[15h] mM
Chapter 4
127
This was in agreement with what observed in the IC50 experiment. In addition, a Ki
value for β-NAD+ of 7.4 ± 1.7 μM was estimated from the Dixon plot (Figure 4.5).
Figure 4.4 IC50 estimation for β-NAD+ toward the activity of NPP from Crotalus adamanteus venom and
15h. Conditions: NPP (0.007 U/mL), 15h (4.4 μM), β-NAD+ (0 – 350 μM), MgCl2 (10 mM), Tris/HCl
buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final
concentrations. Initial velocity values (V0) were calculated from the corresponding time-courses
measuring the fluorescence signals generated when 10% product was formed. The insert shows the range
of β-NAD+ (0 – 25 μM) where the NPP activity drastically decreases. Each data point is the mean (± SD)
of experiments run in duplicate, data plotting with Excel.
Figure 4.5 Dixon plot for β-NAD+ inhibition toward the activity of NPP from Crotalus adamanteus
venom and 15h. Conditions: NPP (0.007 U/mL), 15h (0.2 – 5 μM), β-NAD+ (8 – 15 μM), MgCl2 (10
mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations
are final concentrations. Initial velocity values (V0) were calculated from the corresponding time-courses
measuring the fluorescence signals generated when 10% product was formed. Each data point is the mean
(± SD) of experiments run in duplicate, data plotting with Grafit5. The x value for the lines intersection is
equal to -Ki; the y value for the lines intersection is equal to 1/Vmax.
0
0.0001
0.0002
0.0003
0.0004
0.0005
0.0006
0.0007
-40 10 60 110 160 210 260 310 360
V0
[mM
/se
c]
[β-NAD+] mM
0
0.0005
0.001
0.0015
0 10 20 30
[b-NAD+] mM
-15 -10 -5 0 5 10 15 20
1/V
0
0
10000
20000
30000
40000
50000
[15h] 0.2uM
[15h] 0.6 uM
[15h] 1uM
[15h] 3uM
[15h] 5uM
Data set #6
Data set #7
Data set #8
Data set #9
Data set #10
[15h]
-Ki
Chapter 4
128
To understand the influence of AMP and β-NMN on the NPP activity and their possible
mode of inhibition related to 15h, kinetic analyses were performed for a range of AMP
(Figure 4.6) and β-NMN concentrations.
Figure 4.6 Dixon plot for AMP inhibition toward the activity of NPP from Crotalus adamanteus venom
and 15h. Conditions: NPP (0.007 U/mL), 15h (0.2 – 5 μM), AMP (8 – 40 μM), MgCl2 (10 mM), Tris/HCl
buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final
concentrations. Initial velocity values (V0) were calculated from the corresponding time-courses
measuring the fluorescence signals generated when 10% product was formed. Each data point is the mean
(± SD) of experiments run in duplicate, data plotting with Grafit5. The x value for the lines intersection is
equal to -Ki; the y value for the lines intersection is equal to 1/Vmax.
From the data analysis, AMP turned to be a competitive inhibitor of 15h; indeed, Vmax
did not change, while the Km value for 15h increased with increasing β-NAD+
concentrations. The estimated Ki for AMP from the Dixon plot was 7.5 ± 1.6 μM.
Oppositely to AMP, b-NMN turned to not compete with 15h for NPP binding site in
our experimental conditions; indeed, both Km and Vmax for 15h toward the NPP activity
remained approximately the same with increasing β-NAD+ concentrations (data not
shown).
Chapter 4
129
4.1.2 Kinetic analysis of NPP activity
The Km and Vmax values obtained for 15h related to the activity of NPP from Crotalus
adamanteus venom were compared to those found in the literature for β-NAD+ and
other fluorescent derivatives (ε-NAD+
, ε-PdAD+ and ε-hy
4PdAD
+) related to several
pyrophosphatases (Table 4.1).89, 223, 226-230
NPPs are generally characterized by a wide
substrate specificity, with the enzyme recognition requiring mainly the presence of a
pyrophosphate bond. However, in detailed studies of some NPPs, particular trends in
the substrate affinities were observed. In particular, a nucleotide pyrophosphatase
isolated from Haemophilus influenzae showed different substrate affinities for the
binding site depending on the aromatic rings of the analysed dinucleotides. It was found
that the enzyme recognition was strongly influenced by the adenine ring and its
modifications, more than by the presence of the nicotinamide ring and its derivatives.
The poor substrate activity of ε-NAD+ appeared to be due more to the steric hindrance
of the modified adenine ring in the active site than to the loss of the exocyclic amino
group and, consequently, H-bonding ability.226
Organism β-NAD+ ε-NAD
+a or ε-PdAD
+b 15h
Km
(mM)
Vmax (μmol·
min-1
·mg-1
)
Km
(mM)
Vmax (μmol·
min-1
·mg-1
)
Km
(mM)
Vmax (μmol·
min-1
·mg-1
)
Haemophilus
parasuis227
0.0041 13.5 0.144
a 2.5
a n.a. n.a.
Haemophilus
influenzae226
0.0054 2.7 0.632
a 3.6
a n.a. n.a.
Yeast228
0.17 0.09 n.a.c n.a. n.a. n.a.
Potato229, 230
0.05 n.a. 0.013a n.a. n.a. n.a.
Rat liver223
0.028 n.a. n.a. n.a. n.a. n.a.
Mouse
splenocytes89
n.a. n.a.
0.29 –
0.33b
n.a. n.a. n.a.
Crotalus
adamanteus n.a. n.a. n.a. n.a. 0.0015 0.001
aε-NAD
+;
bε-PdAD
+;
cnot available.
Table 4.1 Enzymological properties for β-NAD+
, ε-NAD+ and the related dinucleotide ε-PdAD
+, and 15h
with different nucleotide pyrophosphatases.
Chapter 4
130
Compound 15h was found to have a very low Km value related to the NPP from
Crotalus adamanteus venom, much closer to that of the natural β-NAD+ than those of
etheno-derivatives for other NPPs. This result suggested a strong affinity of 15h for the
enzyme active site, due to the similarity of 8-(pyrrol-2-yl) adenine to adenine of the
natural β-NAD+. However, the low Vmax indicated that the substitution in position 8 on
the adenine ring, although tolerated by NPP during the binding of 15h in the active site,
was strongly influencing its catalytic mechanism.
From the IC50 determination and the kinetic analyses performed to understand the β-
NAD+ mode of inhibition, β-NAD
+ was confirmed
to be a competitive inhibitor related
to 15h, indicating their binding on the same ligand site.
Regarding AMP and b-NMN, it would be expected that both, as reaction products, are
competitive inhibitors related to β-NAD+ for the NPP activity. In fact, although AMP
was found to be a strong competitive inhibitor for several NPPs (Table 4.2), β-NMN
had no relevant activity, indicating again a stronger affinity of NPPs for the adenine
moiety of the dinucleotides.226, 227, 230
In accordance with the literature, in this study
AMP turned to be a competitive inhibitor related to 15h, with a Ki value similar to that
of β-NAD+, indicating the strong affinity of NPPs for the adenine moiety of
dinucleotides. In contrast, β-NMN did not exert any inhibition toward NPP from
Crotalus adamanteus venom.
Organism AMP β-NMN
Ki (mM) Ki (mM)
Haemophilus parasuis227
0.0366 inactive
Haemophilus influenzae226
0.0151 inactive
Potato230
0.145 inactive
Crotalus adamanteus 0.0075 inactive
Table 4.2 Competitive inhibition of AMP and β-NMN toward different nucleotide pyrophosphatases.
In addition, kinetic experiments were performed with the other 8-aryl/heteroaryl NAD+
derivatives synthesized in Chapter 2, to confirm the reasonable choice of 15h as a
fluorescent probe for the development of fluorescence-based assays. 15b and 15c acted
both as NNSs toward NPP from Crotalus adamanteus venom, as confirmed by a
Chapter 4
131
fluorescence increase occurring during the enzymatic reaction with NPP (Appendix 8,
Figure A5, Figure A6). However, the fluorescence difference generated along the
reaction course was smaller than that occurring with 15h, because of a smaller
fluorescence internal quenching in 15b and 15c compared to 15h. By contrast, 15e and
15g did not show any fluorescence increase during the progress of the enzymatic
reaction. Assuming their activity as NNSs of NPP because of their belonging to the
class of 8-substituted NAD+ derivatives, this result was predictable for 15e, since no
fluorescence internal quenching was observed compared to its corresponding AMP
derivative, 9e. However, this result was surprising for 15g, since it had shown a large
fluorescence internal quenching compared to its corresponding AMP derivative, 9g.
Probably, because of their extremely low values, the fluorescence assay was not
sensitive enough to detect the fluorescence changes.
4.2 NAD+-glycohydrolases (NADases) and ADP-ribosyl cyclases (ADPRCs)
Ca2+
is a secondary messenger involved in cellular signalling and several intracellular
controlled events.231
The Ca2+
concentration inside the cells is ~ 10 – 100 nM and it is
kept constant by the presence of Ca2+
channels, that actively pump Ca2+
from the
cytosol to the extracellular space or into the endoplasmic reticulum in which it is stored.
Sudden increases in the intracellular Ca2+
concentration, due to its release from the
endoplasmic reticulum, are mediated by different Ca2+
-release pathways. The best
characterized is the IP3 (Inositol triphosphate)-mediated pathway, where IP3 binds and
opens Ca2+
channels on the endoplasmic reticulum, allowing Ca2+
release in the cytosol.
Another less understood pathway is that of Ca2+
-induced Ca2+
-release (CICR), where an
intracellular Ca2+
increase, due to influx from the extracellular space or to the IP3-
pathway, generates a further Ca2+
release from the endoplasmic reticulum via ryanodine
receptors, another type of internal Ca2+
channel, completely independent from the IP3
channel.
Cyclic ADPR, cADPR, has been found to be an endogenous regulator of the CICR
pathway, effective as much as IP3 in the IP3-pathway.232, 233
cADPR is formed by the
removal of NaM from β-NAD+, and cyclization of the remaining ADPR between N-1 of
the adenine ring and C-1′′ of the “northern” ribose (the one linked to NaM in the whole
Chapter 4
132
β-NAD+) in the β-configuration.
234, 235 cADPR metabolism is controlled by different
enzymes, its formation being catalyzed by ADP-ribosyl cyclases, while its degradation
to ADPR catalyzed by cADPR-hydrolases.
ADP-ribosyl cyclase (ADPRC) was first isolated from sea urchin eggs as a membrane-
bound protein, but was then found in higher quantity on the Aplysia invertebrates as a
soluble protein.236
This enzyme initially showed only cyclase activity, catalyzing the
consumption of β-NAD+ to cADPR. In contrast, cADPR-hydrolases were found
prevalently in brain extracts as membrane-bound proteins.237
Homology studies
underlined the similarity of the ADPRC sequence with that of the human antigen CD38,
supporting the idea of them belonging to the same class of enzymes. Nevertheless, a
differentiation in preferential activities has occurred during the evolution of these
enzymes: ADPRC has preferentially developed the cyclase activity, while CD38 shows
prevalently NAD+-glycohydrolase (NADase) activity.
56, 63 Later, ADPRC was shown to
have cADPR-hydrolase activity238
and NADase activity71
when stressed in particular
conditions. Likewise, CD38 was shown to have cyclase activity in conditions that do
not favour the hydrolysis of β-NAD+ to ADPR.
232, 239 Moreover, cADPR-hydrolases
found in brain extracts as membrane-bound proteins,237
showed in turn cyclase and
NADase activity,240
as well as proclaimed NADases showed cyclase activity.241
These enzymes are therefore multifunctional and, depending on the substrate or the
reaction conditions, show preferential activities. A unifying mechanism for Aplysia
ADPRC, CD38 and NADases has been proposed to explain the activity differentiation
between these multifunctional enzymes belonging to the same class (Scheme 4.2). A
partitioning mechanism would be involved, rather than a sequential one, in which a
common oxocarbenium ion would be formed from β-NAD+ after the removal of NaM;
the oxocarbenium ion would then evolve preferentially to cADPR or ADPR by
competitive pathways.238
As will be discussed in 4.2.3, the cADPR-hydrolase activity of
Aplysia ADPRC is an example of a limiting form of the partitioning mechanism, in
which the enzyme sequentially forms cADPR, then hydrolysing it to ADPR in presence
of high enzyme concentrations.
Chapter 4
133
Scheme 4.2 Partitioning mechanism for Aplysia ADPRC, CD38 and NADases.
4.2.1 Fluorescence-based assay for NAD+-glycohydrolase
Compound 15h was tested as a NNS of the commercially available NAD+-
glycohydrolase (NADase) from porcine brain. This enzyme catalyzes the cleavage of
the nicotinamide N-glycosidic bond of β-NAD+ to produce NaM and ADPR.
The enzymatic reaction was monitored by HPLC to check the suitability of 15h as an
NADase substrate. In fact, the enzyme recognised the NNS and consumed it catalyzing
its hydrolysis at the nicotinamide N-glycosidic bond to generate the ADPR derivative,
23, and NaM (Figure 4.7).172
The reaction product, 23, was identified by 1H NMR.
Once it was established that 15h was used as a NNS by NADase, the monitoring of the
NADase activity by fluorescence was attempted. However, NADase is a membrane-
bound enzyme and, consequently, completely insoluble in the reaction buffer. This
problem made it difficult to carry out the enzymatic reaction on a microplate reader,
unless a soluble form of the enzyme was used,87, 91, 242, 243
or samplings during the
experiments were performed.243, 244
Chapter 4
134
Figure 4.7 HPLC time-course of the enzymatic reaction of NADase from porcine brain and 15h.
Conditions: NADase (0.009 U/mL), 15h (24.74 μM), Tris/HCl buffer (50 mM, pH 8), 37 ºC; all
concentrations are final concentrations. Data from single experiment, peak area vs time [min], data
plotting to a 1st order rate equation with Grafit5.
Since the purpose of the experiments was not to develop a fluorescence-based assay for
NADase, but to show the applicability of 15h as a biochemical tool to monitor several
enzymatic reactions, we did not attempt the solubilization of the NADase. Instead, the
fluorescence monitoring of the enzymatic reaction was performed in a cuvette to find
out the suitability of 15h for such applications. Unfortunately, no fluorescence increase
was observed. This result was ascribed to the turbidity of the reaction mixture and the
scattering due to the enzyme in suspension, which did not allow any measurement of
fluorescence changes.
The only alternative to prove a fluorescence change during the progress of the
enzymatic reaction was to sample the reaction mixture and read its fluorescence using
the microplate reader. After filtering, diluting and reading the fluorescence of the
samples, a fluorescence increase was finally observed and a time-course obtained
(Figure 4.8), supporting once again the idea of using 15h as a fluorescent probe for the
monitoring of NAD+-consuming enzyme activities. These results suggested that it is
likely that a continuous fluorimetric assay could be set up with 15h using a solubilized
form of the NADase enzyme.
Chapter 4
135
Figure 4.8 Fluorimetric assay of NADase from porcine brain and 15h. Fluorescence time-course as
fluorescence emission [RFU] vs time [min]. Conditions: NADase (0.0075 U/mL), 15h (5.25 μM),
Tris/HCl buffer (50 mM, pH 8), 37 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final
concentrations. Data from single experiment, data plotting to a 1st order rate equation with Grafit5.
4.2.2 Fluorescence-based assay for ADP-ribosyl cyclase
Compound 15h was tested as a NNS of the commercially available ADP-ribosyl cyclase
(ADPRC) from Aplysia californica. This enzyme catalyzes the cyclization of β-NAD+
to cADPR with removal of NaM.
It is known from the literature that 8-substituted NAD+ derivatives are recognised by
this enzyme and used as NNSs to give 8-substituted cADPR derivatives.112
The
enzymatic reaction was monitored by HPLC to check the suitability of 15h as an
ADPRC substrate. In fact, the enzyme recognised the NNS and consumed it, catalyzing
the removal of NaM and cyclization of the remaining ADPR intermediate to the cADPR
derivative, 24 (Figure 4.9).172
The reaction product, 24, was identified by 1H NMR.
time [min]
0 20 40 60 80 100 120
Flu
ore
sce
nce
Inte
nsity
0
2000
4000
6000
8000
10000
12000
23
Flu
ore
scen
ceEm
issi
on
[RFU
]
Chapter 4
136
Figure 4.9 HPLC time-course of the enzymatic reaction of ADPRC from Aplysia californica and 15h.
Conditions: ADPRC (0.05 U/mL), 15h (12.37 μM), HEPES buffer (50 mM, pH 7.4), 25 °C; all
concentrations are final concentrations. Data from single experiment, peak area vs time [min], data
plotting to a 1st order rate equation with Grafit5.
The possibility to develop a fluorescence-based assay to monitor the ADPRC activity
using 15h as a fluorescent probe was explored. A fluorescence increase during the
enzymatic reaction due to the loss of NaM was expected, by analogy with the behavior
shown with NPP and NADase. However, it was not clear whether a fluorescence
increase or decrease would occur following the cyclization of the ADPR intermediate to
the cADPR derivative, 24.
Several attempts were made in a cuvette to find the right substrate/enzyme ratio which
allowed a fluorescence change to be observed during the progress of the enzymatic
reaction. The results allowed us to visualize the mechanism suggested in the literature
for the enzymatic reaction.26
Indeed, as shown in Figure 4.10, moving to more diluted
enzyme concentrations and increasing the substrate/enzyme ratio (as indicated in the
brackets), the enzymatic reaction became increasingly slower highlighting the existence
of two or more reaction steps.
Chapter 4
137
Figure 4.10 Fluorescence changes observed during the enzymatic reaction between ADPRC from Aplysia
californica and 15h. Conditions: ADPRC (0.000055 – 0.0055 U/mL), 15h (0.09 μM), HEPES buffer (50
mM, pH 7.4), 25 °C, λex = 300 nm, λem = 410 nm; all concentrations are final concentrations. Data
plotting with Excel.
At the moment of enzyme addition to start the reaction, a sudden fluorescence increase
was observed, probably due to the removal of NaM and, consequently, the disruption of
the internal quenching in 15h, now transformed into the more fluorescent ADPR
intermediate. Then, the latter, bound to the enzyme, would start folding to cyclize into
24. This cyclization would cause a change in the electronic delocalization on the
substituted adenine ring, resulting in a final drop of fluorescence with the release of the
product by the enzyme.
A further confirmation to this two-step reaction, consisting in the removal of NaM and
following cyclization of the ADPR intermediate to cADPR, was given by the HPLC
assay. First, it was observed that the rate constant for the formation of NaM, kNaM, was
twice that for the formation of 24, k24 (kNaM 0.3525 min-1
; k24 0.1700 min-1
), indicating
that the formation of the two products was not simultaneous. Moreover, no trace of the
-10
0
10
20
30
40
50
60
70
0 50 100 150 200 250 300 350
Flu
ore
sce
nce
Em
issi
on
time [sec]
0.0055 U/mL (16x)
0.0028 U/mL (32x)
0.000055 U/mL (1636x)
Fluorescence quenching
Fluorescence increase!
Fluorescence quenching
Chapter 4
138
ADPR intermediate appeared in the chromatogram, suggesting the binding of the
intermediate to the enzyme with subsequent release of the cyclized product.
To be able to kinetically characterize 15h related to ADPRC from Aplysia californica,
we used high enzyme concentrations allowing us to monitor the decrease in
fluorescence due to the release of 24. An enzyme concentration range between 0.025
and 2.5 U/mL was investigated to find the appropriate enzyme/substrate ratio that
allowed the calculation of linear initial velocities (Figure 4.11).
Surprisingly, at the highest enzyme concentrations (1.25 – 2.5 U/mL), after the initial
fluorescence drop due to the formation of 24, a fluorescence increase was observed.172
In the literature it is known that ADPRC from Aplysia californica shows a cADPR-
hydrolase activity unmasked only at high enzyme concentration, revealing this enzyme
to be multifunctional in respect of its apparent unique ADPRC activity.238
Using 15h
and its exclusive fluorescent properties, we were able to monitor this secondary activity
of the enzyme transforming the newly formed 24 into the ADPR derivative, 23.172
Figure 4.11 Fluorimetric assay of ADPRC from Aplysia californica and 15h: initial velocity (V0) as
function of ADPRC concentration. Fluorescence time-course as fluorescence emission [RFU] vs time
[sec]. Conditions: ADPRC (0.025 – 2.5 U/mL), 15h (9.6 μM), HEPES buffer (50 mM, pH 7.4), 25 °C, λex
= 300 nm, λem = 410 nm, gain 15%. Positive control: 15h (9.6 μM); negative control: buffer; all
concentrations are final concentrations. Initial velocity values (V0) were calculated from the reaction time
measuring the fluorescence signals generated when 10% product was formed. Initial velocity values were
fitted to a linear equation (insert) with Excel: y = 7.0142x+1.0531, r² = 0.9348. Each data point is the
mean (± SD) of experiments run in duplicate.
0
2000
4000
6000
8000
10000
12000
0 500 1000 1500 2000 2500 3000
Flu
ore
sc
en
ce
Em
iss
ion
[R
FU
]
time [sec]
[ADPRC] 0.025 U/mL
[ADPRC] 0.05 U/mL
[ADPRC] 0.125 U/mL
[ADPRC] 0.25 U/mL
[ADPRC] 0.75 U/mL
[ADPRC] 1.25 U/mL
[ADPRC] 1.75 U/mL
[ADPRC] 2.5 U/mL
pos. control
neg contr
0
10
20
30
0 1 2 3
V0
[RFU
/se
c]
[ADPRC] U/mL
Chapter 4
139
To estimate Km and Vmax of 15h related to the ADPR-cyclase activity of ADPRC from
Aplysia californica, the substrate concentration was varied to generate a saturation curve
(Appendix 8, Figure A7). Unfortunately, the small range of 15h concentration with a
linear fluorescence response was not enough to generate a saturation curve that could be
used by Michaelis-Menten equation to determine accurately the enzymatic parameters
related to 15h. Therefore, a Km > 34 μM (concentration limit with a linear fluorescence
response for 15h) and Km/Vmax of 3.7 min·mg·L-1
, extrapolated by Hanes-Woolf plot,
could be estimated for 15h (Appendix 8, Figure A8).172
Then, the estimation of Km and Vmax of 15h related to the cADPR-hydrolase activity of
ADPRC from Aplysia californica was attempted. However, also in this case, the small
range of 15h concentration with a linear fluorescence response was not enough to
generate a saturation curve that could be used by Michaelis-Menten equation (Appendix
8, Figure A9). This indicated quite high Km for the cADPR-hydrolase activity of
ADPRC from Aplysia californica when using 15h, as confirmed in the literature by
HPLC studies revealing a Km for the natural cADPR of 1.04 ± 0.15 mM and a Vmax of
3.34 ± 1.15 μmol·min-1
·mg-1
.238
4.2.3 Mechanistic aspects of ADPRC activity
In the literature the reaction mechanism of the ADPRC from Aplysia californica is well
described and consistent with the results reported here, obtained by the fluorimetric and
HPLC experiments, that suggest a two-step reaction (Figure 4.12).26
The enzymatic
reaction proceeds via binding of β-NAD+ in the catalytic site; there, E179 interacts by
H-bonding with 2′′, 3′′-OH of the nicotinamide ribose, destabilizing the nicotinamide N-
glycosidic bond and favouring the formation of the oxocarbenium intermediate
following the release of NaM.30
In the catalytic site, the oxocarbenium intermediate is
stabilized on the nicotinamide ribose and the diphosphate group by an extended network
of H-bonds; slowly, the intermediate undergoes minor movements bringing the adenine
ring, previously in anti conformation and stacked with Y81, into a syn conformation
with the adenine ring stacked with W140. At this stage, N-1 of the adenine ring, more
nucleophilic than N-7, points to the electrophilic C-1′′, where the cyclization takes
place.
Chapter 4
140
Figure 4.12 Reaction mechanism for N-1 cyclization of 15h with ADPRC from Aplysia californica.
This mechanism is in agreement with that suggested by the HPLC assay, where the
immediate release of NaM would be followed by a slower release of the product, since
an intermediate is formed which rearranges to give the final cADPR.
As shown in 3.1.2, 8-aryl/heteroaryl substituted NAD+ derivatives, in particular 15h,
have the adenine ring in syn conformation because of the steric hindrance caused by the
substituent in position 8.70
Consequently, if at the moment of binding, 15h has the
adenine in a syn conformation in contrast to the anti in the natural β-NAD+, then, at the
moment of cyclization, 15h already has the right syn conformation with N-1 pointing in
the direction of C-1′′. Since the rate-limiting step in the reaction mechanism seems to be
the cyclization, the comparable Km values for 15h and β-NAD+ related to ADPRC from
Aplysia californica could be explained by their similarities in the cyclization position
(Table 4.3).172
Fluorescence-based assays to study the activity of ADPRC from Aplysia californica
were already present in the literature and based on the fluorescence changes of the well
known ε-NAD+, or other dinucleotides, such as NGD
+, NHD
+ and NXD
+ (see 1.4.3).
69,
90, 232, 239, 245 Conversely from 15h, all these compounds showed an increase of
fluorescence with the progress of the enzymatic reaction. In particular, NGD+, NHD
+
Chapter 4
141
and NXD+ became fluorescent only after their cyclization to cGDPR, cIDPR and
cXDPR. All these dinucleotides were characterized by the cyclization position on N-7,
due to the lack of availability of N-1, as in the case of ε-NAD+, or to the weak
nucleophilicity of the amidic/imidic N-1 compared to N-7 in the case of NGD+, NHD
+
and NXD+ (Figure 4.13). Studies on the fluorescent properties of adenine rings
substituted on N-1 and N-7 suggested that, while the substitution on N-1 did not cause
any fluorescence, substitution on N-7 generated fluorescence;69
this would explain the
reduced fluorescence of 24 compared to 15h, and the strong fluorescence of cGDPR,
cIDPR and cXDPR compared to NGD+, NHD
+ and NXD
+.
Figure 4.13 Cyclic ADP-ribose derivatives obtained by cyclization on N-1 or N-7 with ADPRC from
Aplysia californica.
β-NAD+
15h ε-NAD+
NGD+
NHD+
NXD+
Km [μM] 3969
, 13558
> 34 6.269
1.769
, 15.558
0.569
, 8.858
6.269
Vmax [μmol·min-1
·mg-1
] 50069
~ 20 94.469
36.369
4.669
5069
Table 4.3 Enzymological properties for β-NAD+, 15h, ε-NAD
+ and other dinucleotide substrates of
ADPRC from Aplysia californica.
Chapter 4
142
All the dinucleotides cyclizing in N-7 of the adenine ring turned out to have extremely
low Km values compared to those of β-NAD+ and 15h (Table 4.3), showing stronger
affinities for ADPRC from Aplysia californica than the natural substrate itself. This
behaviour could be explained with the favourable anti conformation of the purine rings
having N-7 pointing in the direction of C-1′′, ready for the cyclization without the need
to rotate as in the case of adenine in β-NAD+.69
As already mentioned in 4.2, Aplysia ADPRC, characterized at a first investigation only
by the cyclase activity,236
turned out to be a cADPR-hydrolase238
as well as a
NADase.63, 71
These activities show up only in particular conditions in which the
enzyme is forced to follow a different from the preferred mechanism. A partitioning
mechanism, more than a sequential mechanism, is responsible for the choice of the
reactions performed by CD38, NADases and ADPRC belonging to the same family (see
4.2, Scheme 4.2).
In the case of the NADase activity of ADPRC, unless the cyclization is completely
blocked by the use of some NAD+ derivatives,
71 the rate for the cyclization is so fast
compared to the solvolysis that the enzyme would follow a macroscopic sequential
mechanism, where β-NAD+ is first converted to cADPR (cyclase activity), which then
undergoes hydrolysis to give ADPR (cADPR-hydrolase activity). This sequential
reaction mechanism can be clearly visualized in the experiments performed using 15h
as a fluorescent NNS for ADPRC,172
where an initial fluorescence drop, indicating the
formation of 24, changes into a fluorescence increase due to the formation of 23 (Figure
4.11).
In other respects, with ε-NAD+ it would not be possible to distinguish between cyclase
and cADPR-hydrolase activities of ADPRC from Aplysia californica, since its
fluorescence would increase after the formation of both cεADPR and ε-ADPR.
Moreover, cεADPR is quite resistant to the hydrolysis90
and, therefore, it would not
probably be susceptible to the cADPR-hydrolase activity.
In principle, the continuous monitoring of the sequential reaction mechanism for the
formation of ADPR from β-NAD+ via cADPR would be possible using NGD
+, NHD
+
and NXD+. Indeed, these compounds should show a fluorescence increase with the
Chapter 4
143
formation of the cyclic product (cyclase activity), followed by a fluorescence decrease
with the formation of the corresponding nucleotide diphosphoribose derivatives
(cADPR-hydrolase activity). Only NHD+ is actually used to fluorimetrically monitor the
cADPR-hydrolase activity,69, 245
since its cyclic derivative, cIDPR, is less hydrolytically
stable than cGDPR69, 232, 239
and cXDPR. Nevertheless, no application of cIDPR for the
monitoring of the recently discovered cADPR-hydrolase activity of ADPRC from
Aplysia californica has been found in the literature.
The study and understanding of ADPRC from Aplysia californica is important from a
biological perspective since this enzyme belongs to the same family of human CD38,
that is involved in the β-NAD+ regulation and Ca
2+ signalling. In addition, it is used for
the chemo-enzymatic synthesis of well known or novel cADPR derivatives, useful as
biochemical tools or potential inhibitors of the Ca2+
-signalling pathway. Hence, it is
essential to know more about this enzyme, its reaction mechanism and its specifity.
The previous results, showing on analytical scale the occurrence of the cyclization by
ADPRC of 15h into the less fluorescent product, 24, were subsequently confirmed by
the preparation of 24 on a synthetic scale allowing its spectroscopic analysis.246
ADPRC
from Aplysia californica was used to cyclize 15h into 24 on a synthetic scale; the
reaction progress was monitored by HPLC and the reaction mixture purified by anion-
exchange chromatography to give the desired product with a 96% yield (Scheme 4.3).246
Scheme 4.3 Synthetic pathway to 8-(pyrrol-2-yl) cADPR, 24. (i) Suzuki-Miyaura cross-coupling; (ii)
morpholidate formation; (iii) coupling with β-NMN (R); (iv) cyclization with ADPRC from Aplysia
californica.
Chapter 4
144
In the literature it has been reported that cADPR is quite stable in basic conditions;234
indeed, the N1-ribosyl bond does not undergo any hydrolysis or rearrangement at pH 10,
and the pyrophosphate bond is hydrolysed only under strong basic condition. However,
cADPR is slightly acid-labile, its half-life for spontaneous hydrolysis at 37 °C is 24 h,
and it can be easily hydrolysed by NADases or ADPRC with cADPR-hydrolase
activity, but not by NPPs.234, 235
Compounds 15h and 24 were subjected to similar
hydrolytic conditions (0.1 N NaOH and HCl) to identify the chemical stability of 24
compared to 15h (Appendix 8, Figure A10).
Spectroscopic and 1H NMR analyses for 24 were performed and compared to 15h; the
results are summarised in Table 4.4 and Table 4.5.246
Compound 24 showed a
substantial red shift in λex compared to 15h, with consequent decrease in Stoke shift,
which indicated a minor vibrational relaxation due to the constrained conformation of
the adenine ring cyclized in N-1 position. The value of 24, as expected from the
fluorescence-based assay for ADPRC, was less than 15h, probably due to the change of
the electron delocalization along the adenine ring, now positively charged.
The 1H NMR spectrum clearly showed the cyclization product. Indeed, the nicotinamide
signals disappeared, while a considerable downfield shift was noticed for 2-H on the
adenine ring and, consequently, for 1′-H and 1′′-H, deshielded because of the newly
formed N1-ribosyl bond and the positively charged adenine ring.
cmpd λex/nm λem/nm Stoke shifta/cm
-1
15h 309 374 5624 0.0053
24 316 378 5190 0.004 a Stokes shift = (1/ex - 1/em).
Table 4.4 Spectroscopic analysis and values of 15h and 24 in water.
cmpd 2-H (ppm) 1′-H (ppm) 1′′-H (ppm)
15h 8.09 5.73 5.91
24 8.98 6.42 6.16
All the spectra are referenced for D2O ( 4.79 ppm).
Table 4.5 1H NMR analysis for 15h and 24. Comparison of chemical shift (ppm) for 2-H, 1′-H and 1′′-H.
Chapter 5
145
5 EVALUATION OF ARYL/HETEROARYL ADENINE-MODIFIED NAD+ DERIVATIVES AS INHIBITORS AND NON-NATURAL SUBSTRATES FOR NAD+-CONSUMING ENZYMES
In Chapter 4, 15h, the most promising fluorescent NAD+ derivative belonging to the
class of those 8-aryl/heteroaryl substituted, was evaluated as a biochemical probe for
NAD+-consuming enzymes; in particular, its role as a NNS was analysed. Of course, the
activity of 15h as a NNS for NAD+-consuming enzymes implies a competitive
inhibition with the natural substrate β-NAD+, which was only superficially explored in
the previous enzymatic evaluation of 15h.
Herein, greater attention will be given to the inhibitory effects of the aryl/heteroaryl
adenine-modified NAD+ derivatives toward NAD
+-consuming enzymes. It will be
stressed how competitive inhibition does not necessarily correlate with substrate
activity, as was the case with the enzymes analysed in Chapter 4. Instead, for complex
enzymes that involve more than one substrate in their reactions, the situation can change
due to the more complicated catalytic mechanisms and active sites, requiring elaborate
and discriminating factors in the recognition and use of their substrates. NAD+-
dependent DNA ligases and sirtuins have been chosen in this project as complex NAD+-
consuming enzymes to evaluate the biological activities of the differently substituted
NAD+ derivatives as inhibitors and/or NNSs. These enzymes require β-NAD
+ as a co-
substrate in their reactions and have in their structures specific adenine binding
domains, which reflect on the different activities exerted by the adenine-modified NAD+
derivatives as inhibitors and/or NNSs.
5.1 NAD+-dependent DNA ligases
DNA is continuously exposed to damage and lesions deriving from the physiological
processes in which it is involved, for instance, mismatches during DNA replication, or
DNA strand breaks due to aborted topoisomerase activity. However, DNA damage is
mostly due to processes unrelated to its biological roles, like the attack from ROS and
environmental factors, such as ultraviolet light, radioactive exposure or tobacco
products.
Chapter 5
146
To defend themselves, living cells have developed complicated and efficient systems to
detect DNA damage and repair it; the DNA-damage response (DDR) implies an
intricate cascade of signalling processes and enzymatic activities that are not yet fully
understood, but are continuously under study.247
DNA polymerases, ligases,
methyltransferases, helicases, endonucleases, PARPs and many other enzymes are
fundamental components of the DDR. Defects in DDR are associated with the
development of many diseases, such as heritable neurodegenerative diseases and cancer.
In the latter case, there is a strong connection between cancer and DDR, since defects in
DDR cause genome instability and, therefore, mutations and malignancies; however,
DDR is also responsible for cancer-therapy resistance. Consequently, DDR inhibitors
can in some cases be used for anticancer therapy.
DNA ligases are essential enzymes found in all living cells, involved in DNA
replication, repair and recombination;15, 248, 249
their activity consists of sealing DNA
single-strand breaks between 5′-phosphate and 3′-hydroxyl termini. DNA ligases have
been differentiated into two classes depending on their co-substrate specificity;
eukaryotic DNA ligases are ATP-dependent, while bacterial and some viral DNA
ligases are NAD+-dependent. In fact, ATP-dependent DNA ligase sequences have been
found in bacteria in addition to the NAD+-dependent ones, even though they are
expressed only in particular conditions;250
in contrast, NAD+-dependent DNA ligases
have never been found in eukaryotic organisms.
The restricted distribution in bacteria, the necessary requirement for bacteria viability
and the absolute co-substrate uniqueness make NAD+-dependent DNA ligases a
possible target for the development of antibacterial drugs. NAD+-dependent DNA
ligases from different bacteria have been structurally characterized to better understand
their activities and reaction mechanism, and to develop possible inhibitors. High
homologies have been found between sequences of several bacterial DNA ligases,
showing highly conserved motifs and the same modular architecture. The sequence
similarity with ATP-dependent DNA ligases is quite poor, as expected because of the
different co-substrate; however, NAD+-dependent and ATP-dependent DNA ligases
share exactly the same reaction mechanism (Scheme 5.1).249
Chapter 5
147
Scheme 5.1 Reaction mechanism for DNA ligation by ATP and NAD+-dependent DNA ligases.
The ligation of DNA occurs in three steps. In the first step, ATP or β-NAD+ binds to the
active site of the enzyme; there, a nucleophilic lysine residue attacks the phosphate of
the adenosine in the co-substrate with subsequent release of the leaving group,
respectively pyrophosphate (PPi) and β-NMN, and formation of adenylated-ligase
intermediate (1). In the second step, after the positioning of the DNA nick in the
Chapter 5
148
proximity of the adenylated-ligase intermediate (1), the nucleophilic phosphate in the 5′
position attacks the phosphoramidate bond in 1 to form an adenylated-DNA
intermediate (2). In the third and final step, the nucleophilic hydroxyl group in the 3′
position attacks the new pyrophosphate bond with formation of a phosphodiester bond
between the 5′ and 3′ positions of the DNA (3), and release of AMP.
The NAD+-dependent DNA ligases have a modular architecture consisting of four main
domains linked by single chains (Figure 5.1). Domain 1 is formed by two subdomains;
subdomain 1a, at the N-terminus, binds the β-NMN moiety of β-NAD+, while
subdomain 1b is the adenylation domain, or the nucleotidyltransferase domain (NTase),
since it binds the AMP moiety of β-NAD+. Domain 2 is the oligonucleotide binding site
(OB). Domain 3 consists of subdomain 3a, a Cys4-type zinc finger, and subdomain 3b,
containing a helix-hairpin-helix motif (HhH). Domain 4, at the C-terminus, is a BRCT
domain (Breast cancer Carboxy-Terminal). The catalytic core of DNA ligase consists of
subdomain 1b and OB, flanked by N and C-terminal domains arranged in a C-shape
that clamps the double-stranded DNA.251
Subdomain 1a has been shown to interact with the β-NMN moiety of β-NAD+, indeed it
is not present in ATP-dependent DNA ligases.252
Its presence is fundamental to initiate
the ligation reaction, but not essential in the case of preadenylated-enzyme and its
interaction with DNA; this means that subdomain 1a is required for the β-NAD+ binding
and the adenylation step. From crystallographic studies, it was possible to identify an
“open” enzyme conformation, in which β-NAD+ was bound to the solvent exposed
NTase domain. Therefore, the closure of subdomain 1a to bury β-NAD+ and interact
closely with β-NMN puts the latter in an apical position as a leaving group during the
nucleophilic attack of the lysine residue on the adenosine phosphate; the subdomain 1a
has to swivel by almost 180° to switch from the “open” to the “closed” conformation.253
Chapter 5
149
Figure 5.1 Structure of DNA ligase from E. coli (EcLigA) (PDB 2OWO) in complex with nicked DNA.
Visualization by PyMol software. Subdomain 1a (light blue), subdomain 1b (green), OB domain
(yellow), subdomain 3a (orange), subdomain 3b (red).
Subdomain 1b is the adenylation domain, or the core of the DNA ligase. Several
residues have been found to be fundamental in the NTase for the enzyme activity, their
modifications bringing a considerable rate decrease or inactivation of the enzyme. The
main residues in NTase, their interactions and their roles are summarised in Table
5.1.252, 254
The OB domain is involved in the oligonucleotide binding, as it contacts the double-
strand DNA covering the nick and help in positioning it into the NTase domain. Its
action is supported by subdomain 3b, HhH, which binds both DNA strands in a non-
specific sequence manner across the minor groove. OB, NTase and HhH clamp the
double-strand DNA along 19 bp.255
1b1a
3b (HhH)
3b (Zn)
2 (OB)
Chapter 5
150
Residue Pos. Interaction Role
Lysine 115 Part of KXDG conserved motif; nucleophilic
attack to adenosine phosphate of β-NAD+
Formation of adenylated-
ligase intermediate (1)
Aspartate 117 Part of KXDG conserved motif Stabilization active site
Glycine 118 Part of KXDG conserved motif Stabilization active site
Glutamate 173 Coordination of O-2′ of the adenosine ribose in β-
NAD+ and AMP
Involvement in step 1 and 3
Lysine 290 H-bonding with N-1 on adenine ring Stabilization of adenine
conformation
Lysine 314 H-bonding with the adenosine phosphate and O-2′
of the nicotinamide ribose in β-NAD+
Involvement in step 1 and 3
Arginine 136 H-bonding with β-NAD+ pyrophosphate and 5′-
phosphate side of the nick
Involvement in step 1 and 2
Arginine 200 Interaction with subdomain 1a in closed
conformation of adenylation domain; interactions
with 3′-OH side of nick and DNA minor groove
Involvement in step 1 and 3
Arginine 208
Arginine 218 Interaction with 3′-OH side of the nick Involvement in step 2 and 3
Arginine 277 H-bonding with Ile284 Stabilization active site
Aspartate 285 Interaction with Mg2+
Metal binding
Arginine 308 Interaction with 5′-phosphate side of the nick Involvement in step 2 and 3
Table 5.1 Essential residues in the NTase domain of EcLigA and their roles.
Subdomain 3a, the Cys4-type zinc finger, seems to hold OB and HhH during DNA
binding, showing therefore a structural role.251, 255, 256
Finally, the C-terminal domain 4,
BRCT, has a controversial role in DNA ligases; indeed, from the crystal structures of
different enzymes, BRCT appears as a very flexible unit, with a structure that cannot to
be properly resolved, and it is apparently not involved in the adenylation step. Indeed,
DNA ligase proteins not including BRCT in their modular structures are still able to
perform their ligation reaction, despite a lower rate. Moreover, DNA binding was
shown to be tighter for a DNA ligase with BRCT domain compared to DNA ligase not
containing it. Therefore, the BRCT domain could stabilize the DNA binding on DNA
ligases, even though it does not interact directly with DNA.251, 255-257
Another possible
Chapter 5
151
function as a signal transducer of the DNA-damage response could be assigned to
BRCT; indeed, other well known proteins use this domain for protein-protein
interactions to transmit signalling responses.251
Overall, the modular structure of NAD+-
dependent DNA ligase allows it to catalyze the ligation reaction involving considerable
movements of the different domains.
The first DNA ligase inhibitors were found by screening libraries of classes of
compounds, including alkaloids, flavonoids, pyridochromanones, quinoline and
quinacrine derivatives, and many others.248, 258-260
The scaffolds of the compounds
showing inhibitory activity were subsequently optimized on the basis of computational
studies to improve activity, selectivity, specificity, solubility, and many other factors.73,
74, 261, 262 However, a deeper knowledge of the structure of NAD
+-dependent DNA
ligases allowed the design of more specific inhibitors. At the outset of this project, the
crystal structures of several NAD+-dependent DNA ligases were resolved and
published, showing the existence of a hydrophobic tunnel between the outside and the
adenylation domain next to N-1, C-2 and N-3 of the adenine ring of β-NAD+; in
particular, C-2 pointed exactly in the direction of the tunnel (Figure 5.2).15, 255
Figure 5.2 Hydrophobic tunnel on NTase domain of DNA ligase from Enterococcus faecalis (PDB
1TAE). C-2 of the adenine ring points in the direction of the tunnel. Visualization by PyMol software.
Chapter 5
152
Many known inhibitors were docked into the active site and shown to fit into this
tunnel.15
This observation initiated the discovery of NAD+-dependent DNA ligase
inhibitors based on mimics of the adenine/adenosine structure substituted in position
2,261, 262
which can compete for β-NAD+ binding. Indeed, ATP-dependent DNA ligases
do not have this tunnel in their structure, providing the possibility that such inhibitors
may act specifically against NAD+-dependent DNA ligases.
Until now, different adenosine/ADP/ATP derivatives substituted in position 2 with alkyl
or cycloalkyl substituents have been tested in vitro/in vivo as antibacterial agents with
promising results (Figure 5.3).73, 74, 263
However, to the best of our knowledge no AMP
or NAD+ derivatives have been reported as inhibitors of NAD
+-dependent DNA ligases.
It is likely that the difficult synthesis of NAD+ derivatives, their hydrophilicity, their
flexible/multifunctional structure making β-NAD+ binding quite unspecific, together
with their potential NNS activity interfering with that as inhibitors, have been the
reasons why such compounds have not been described yet.
Figure 5.3 C-2 adenine-modified adenosine derivatives and their efficacies as NAD+-dependent DNA
ligases inhibitors.73, 74, 263
In this project, the biological activity of the NAD+ derivatives and some of their
corresponding AMP and AMP-morpholidate derivatives, modified in position 2, 6 or 8
of the adenine ring, as inhibitors and/or NNSs was assessed toward EcLigA, one of the
best characterized NAD+-dependent DNA ligases. In particular, the whole β-NAD
+
scaffold, differently from adenosine or AMP scaffolds that could bind also to ATP-
dependent DNA ligases, should provide selectivity toward NAD+-dependent DNA
ligases, as well as the introduction of substituents in position 2 on the adenine ring. In
Chapter 5
153
addition, the introduction of aryl/heteroaryl substituents in position 2, 6 or 8 of the
adenine ring and the biological activities of the corresponding derivatives would
improve understanding of the enzyme active site and the essential interactions for the β-
NAD+ binding.
5.1.1 Expression, purification and activity of Escherichia coli DNA ligase
Escherichia coli DNA ligase (EcLigA) is a protein consisting of 671 amino acids, with
a molecular weight of 74 kDa. The protocol for the overexpression and purification of a
recombinant form of EcLigA, containing a 10-His-tag at the N-terminus of the protein
that allows its straightforward purification, is well established in the literature257
and it
was followed with minor changes to obtain highly pure EcLigA with 24 μM
concentration (Figure 5.4).
Figure 5.4 Purification of EcLigA. Lane M: marker of protein molecular weights; lane 1: crude extract of
overexpressed proteins in the soluble fraction; lane 2: 10 x dilution of purified protein; lane 3: purified
protein. Visualization of SDS-PAGE gel by Coomassie blue staining.
The activity of EcLigA was tested by a gel electrophoresis assay of the ligation of
nicked DNA (Figure 5.5 a). The DNA substrate was prepared by annealing of a 40-mer
strand (LigSub1+2) with two complementary 18-mer (LigSub1) and 22-mer (LigSub2)
strands forming a nick between the 18th
and 19th
bases of one strand; LigSub1 and
LigSub2 were characterized by the addition of fluorescein (F) and a phosphate group
(P) in their 5′ positions respectively. For the preparation of this substrate, the
74 kDa83 kDa
62 kDa
47.5 kDa
1 2 3M
Chapter 5
154
appropriate ratio between the different strands and salt concentration were established to
obtain complete annealing (Appendix 8, Figure A11). The ligation reaction was
monitored by formation of a fluorescent 40-mer strand, migrating less than the
fluorescent 18-mer strand on a denaturing gel, which separates DNA strands based on
their molecular size (Figure 5.5 b).
Figure 5.5 Ligation assay with fluorescent annealed nicked substrate. a) Scheme of ligation assay with
fluorescent annealed nicked substrate; b) denaturing PAGE for time-course of ligation reaction with
fluorescent annealed nicked substrate. Conditions: DNA substrate (0.125 μM), EcLigA (0.1 μM), β-
NAD+ (26 μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all
concentrations are final concentrations. Reactions incubated at 30 °C under shaking; samplings at 0, 1, 2,
3, 5, 10, 15 and 30 min. Gel visualization by fluorescence imaging.
Preliminary experiments were carried out to establish the conditions for the ligation
reaction during the inhibition and NNS activity studies. Indeed, it is important to choose
the right enzyme concentration and, consequently, the time sampling, to have 40 – 50%
ligation in the control reaction without inhibitor. These conditions, varying from batch
to batch of enzyme, allow clear detection of inhibition and activation of the enzyme
activity in the same set of experiments, since the ligation is still not complete and not
influenced by displacement of the inhibitors by the natural β-NAD+. Therefore, the
following studies investigated the reaction in an initial state, not influenced by
reversible reactions, inhibition by products or enzyme deactivation.
For the NNS activity test, the reactions in some cases were maintained for longer to
allow the NNS to be recognised and metabolized by the enzyme. Therefore, the enzyme
activity and the substrate stability were checked over time to be sure they would not
negatively influence the ligation reactions. EcLigA retained full activity after 90 min at
fluor5′
OH P
3′
3′ 5′
ligationfluor
LigSub1 LigSub2
LigSub1+2
a b
0’ 1’ 2’ 3’ 5’ 10’ 15’ 30’
Chapter 5
155
30 °C; however, the activity dropped dramatically after 3.5 h at 30 °C. The annealed
substrate was shown to be stable over 1 h (data not shown).
To validate the gel electrophoresis assay and its application in the inhibition studies,
quinacrine, a well characterized inhibitor of EcLigA, was tested at different
concentrations and its IC50 estimated and compared with the value in the literature
(Appendix 8, Figure A12). The obtained value, 3.9 ± 1.5 μM, was close to that in the
literature, 1.5 ± 0.2 μM,260
indicating the suitability of the method in studying inhibitors.
At this stage, it was possible to proceed with the biological testing of the synthesized
compounds. The series of NAD+ derivatives substituted on either C-2, C-6 or C-8 and
some of their corresponding AMP and AMP-morpholidate derivatives were
successively tested as NNSs and inhibitors of EcLigA.
5.1.2 Biological testing of 2-substituted AMP and NAD+ derivatives
Firstly, 2-substituted AMP and NAD+ derivatives, respectively 7a-b and 13a-b, were
tested as NNSs of EcLigA at 250 μM in the absence of β-NAD+ (Figure 5.6); this
concentration, 10 fold larger than that used for the assay with the natural β-NAD+, was
chosen to ensure detection of any substrate activity. From these experiments, a NNS
activity was observed for both the NAD+ derivatives, 13a and 13b, but not for the AMP
derivatives, 7a and 7b. This was expected because the AMP derivatives have a
phosphate that is not reactive enough to be attacked by the nucleophilic lysine residue in
the active site (see 5.1, Scheme 5.1).
When the NNS activity for 13a and 13b was tested at different substrate concentrations
between 50 and 250 μM, 13a showed the same or a lower activity than the natural β-
NAD+ at 26 μM (Figure 5.7 a), while 13b showed the same or slightly higher activity
than the natural β-NAD+ at 26 μM (Figure 5.7 b).
Chapter 5
156
Figure 5.6 Denaturing PAGE for the NNS activity of 2-substituted AMP and NAD+ derivatives at 250
μM. Conditions: DNA substrate (0.5 μM), EcLigA (0.05 μM), 2-substituted derivative (250 μM) in 1 x
buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations are final
concentrations. Reactions incubated at 30 °C under shaking and sampled at 15 min. Control reaction at
the same conditions with β-NAD+ (26 μM) without any 2-substituted derivative. Gel visualization by
fluorescence imaging.
Figure 5.7 NNS activity of 2-substituted NAD+ derivatives at concentrations between 50 and 250 μM.
Conditions: DNA substrate (0.5 μM), EcLigA (0.05 μM), 13a (a) (50 – 250 μM), 13b (b) (50 – 250 μM)
in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations are
final concentrations. Reactions incubated at 30 °C under shaking and sampled at 5 min. Control reaction
at the same conditions with β-NAD+ (26 μM) without any 2-substituted derivative. Analysis of assay run
on denaturing PAGE and gel visualized by fluorescence imaging. Data analysed using ImageJ. Each data
point is the mean (± SD) of a set of experiments run in triplicate; data plotting with Excel.
Next, inhibition studies were undertaken testing 2-substituted AMP/AMP-
morpholidate/NAD+ derivatives, respectively 7a-b, 10b, and 13a-b, at 200 μM with β-
NAD+ at 26 μM; the samplings were performed at 5 and 15 min (Figure 5.8 and
Appendix 8, Figure A13). A clear inhibition was observed for the 2-iodo AMP and
NAD+ derivatives, respectively 7a and 13a, which showed around 80% inhibition at 200
μM at 5 min. Regarding the 2-phenyl derivatives, a clear inhibition under the
experimental conditions was observed only for the AMP-morpholidate derivative, 10b,
0
20
40
60
80
100
control 50 100 150 200 250
% li
gati
on
0
20
40
60
80
100
control 50 100 150 200 250
% li
gati
on
a b
Chapter 5
157
showing around 70% inhibition, while AMP and NAD+ derivatives, respectively 7b and
13b, showed no inhibition. At 15 min, the ligation for all reactions reached the level of
the control reaction. IC50 values were measured for 7a and 13a (Appendix 8, Figure
A14), and are summarised in Table 5.2.
Figure 5.8 Inhibition of 2-substituted AMP/AMP-morpholidate/NAD+ derivatives at 200 μM. Conditions:
DNA substrate (0.5 μM), EcLigA (0.05 μM), β-NAD+ (26 μM), 2-substituted derivative (200 μM) in 1 x
buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations are final
concentrations. Reactions incubated at 30 °C under shaking and sampled at 5 and 15 min. Control
reaction at the same conditions without any 2-substituted derivative. Analysis of assay run on denaturing
PAGE and gel visualized by fluorescence imaging. Data analysed using ImageJ. Each data point is the
mean (± SD) of a set of experiments run in triplicate; data plotting with Excel.
cmpd scaffold Position of substitution R (substituent) IC50 (μM)
7a AMP 2 Iodo 119.9 ± 23.4
13a NAD 2 Iodo 137.6 ± 17.8
2-MeSADP263
ADP 2 Methyl 4.1 ± 0.5
2-MeSATP263
ATP 2 Methyl 1.1 ± 0.4
Table 5.2 IC50 values ± SD for 7a and 13a toward EcLigA.
An activity comparison between 2-iodo and 2-phenyl substituted derivatives was
undertaken to investigate the influence of the substituents on the different biological
activities observed for the compounds. EcLigA was inhibited by 2-iodo NAD+, 13a
(Figure 5.8); however, due to its NNS activity (Figure 5.7 a), it actively competed with
β-NAD+ for the same active site and catalytic reaction. Since in the inhibition study 13a
was used in a large excess (200 μM) compared to β-NAD+ (26 μM), EcLigA (0.05 μM)
would have bound in its active site more 13a than β-NAD+, independently from the
different affinities (Km) of the compounds (Figure 5.9). The fact that the resulting
0
20
40
60
80
100
control 7a 13a 7b 13b 10b
% li
gati
on
5 min
15 min
Chapter 5
158
ligation was much slower than the control reaction with only β-NAD+ and gave at 5 min
an inhibition overall, indicated that 13a was used preferentially as substrate because of
its excess in the reaction media, but it had a very slow turnover (low kcat) along the
different reaction steps compared to β-NAD+. Indeed, at 15 min the ligation reached the
same level of the control reaction, strongly suggesting a kinetic difference between the
reaction steps involving 13a and β-NAD+.
In contrast to 13a, 2-phenyl NAD+, 13b, seemed to have no inhibitory activity toward
EcLigA (Figure 5.8); however, its NNS activity (Figure 5.7 b) implied a competition of
13b with β-NAD+ for the same active site. Also in this case, 13b was used in a large
excess (200 μM) compared to β-NAD+ (26 μM); therefore, EcLigA (0.05 μM) would
have bound in its active site more 13b than β-NAD+, independently from the different
affinities (Km) of the compounds (Figure 5.9). The fact that the resulting ligation at 5
min was the same as that for the control reaction, suggested that 13b had a very similar
turnover compared to β-NAD+, hence EcLigA could perform its ligation at a similar rate
than with β-NAD+.
The 2-substituted AMP derivatives, 7a and 7b, as well as the corresponding NAD+
derivatives, 13a and 13b, showed a distinct activity depending on the substituent
introduced in position 2 on the adenine ring. Owing to the lack of the NNS activity due
to the AMP scaffold, 7a and 7b provided some clues about the different affinities of the
compounds, generated by the different substituents, toward the active site of EcLigA.
The 2-iodo AMP, 7a, acted as an inhibitor of EcLigA in the presence of β-NAD+ at 5
min (Figure 5.8); because of its analogy with 13a, 7a would presumably bind in the
same β-NAD+ binding site, and the inhibitor activity would indicate its good affinity for
EcLigA. The increased ligation at 15 min would be due to the displacement of 7a by β-
NAD+
with the re-establishment of the normal enzyme activity. In contrast to 7a, the 2-
phenyl AMP, 7b, showed no inhibition (Figure 5.8). This effect could be due to a low
affinity of 7b toward EcLigA; hence, 7b would not bind, or just weakly, to EcLigA and
could be easily displaced by β-NAD+, giving no substantial inhibition overall.
Chapter 5
159
Finally, the inhibitor activity of the 2-phenyl AMP-morpholidate, 10b, (Figure 5.8)
would be due to its modification on the phosphate, since its corresponding AMP
derivative, 7b, showed no inhibitor activity at all.
Figure 5.9 Competition of 2-substituted NAD+ derivatives, 13a-b (light blue), with β-NAD
+ (green) for
EcLigA (E, brown). k1 is the binding constant of β-NAD+ to EcLigA; k2a and k2b are the binding constants
respectively of 13a and 13b to EcLigA; k3 is the rate constant for the displacement reaction; k4 and k4a/b
are the rate constants for the enzyme adenylation (E·AMP*) respectively with β-NAD
+, 13a and 13b; k5
and k5a/b are the rate constants for the nicked DNA adenylation (E + AMP*·nDNA) respectively with β-
NAD+, 13a and 13b; k6 and k6a/b are the rate constants for the ligation (E + AMP + DNA) respectively
with β-NAD+, 13a and 13b. In the inhibition experiments, where the excess of 13a-b shifts the equilibria
toward the binding of 13a-b compared to β-NAD+, the resulting ligation is due to differences in the rate
of any of the steps involved in the metabolism of β-NAD+ and 13a-b.
E
E E
E E
E + NAD+
E · AMP*
E + AMP* · nDNA
E + AMP+ DNA
K1 K2a/b
K3
K4 K4a/b
K5a/bK5
K6 K6a/b
Chapter 5
160
5.1.2.1 Biological activities of 2-substituted AMP and NAD+ derivatives
In agreement with the already developed inhibitors based on modifications of C-2 on
the adenine ring of adenosine, ADP or ATP structures (see 5.1, Figure 5.3),73, 74, 263
2-
substituted AMP/AMP-morpholidate/NAD+ derivatives were found to be biologically
active as inhibitors and/or NNSs of EcLigA, suggesting the validity of the introduction
of substituents in position 2 on the adenine ring.
Interesting observations came out of the screening of the activities of the different
compounds. As suggested by the combined observations of NNS and inhibitor
activities, 2-iodo NAD+, 13a, was a NNS for EcLigA with a slow turnover compared to
β-NAD+, causing an inhibitory effect overall. In addition, 2-iodo AMP, 7a, not a NNS
because of its AMP scaffold, showed inhibitor activity and, therefore, a certain affinity
for EcLigA. The activities for 2-phenyl NAD+, 13b, and 2-phenyl AMP, 7b, were less
clear. At first glance, the results showing no inhibition could be easily interpreted as no
or weak binding to EcLigA; however, the NNS activity of 13b suggested a similar
binding and turnover compared to β-NAD+, hence justifying the same extent of ligation
as the control reaction. In contrast, 7b, due to the AMP scaffold, could not be a NNS for
EcLigA, but neither did it not show any inhibitory activity, therefore suggesting that 7b
could indeed have no activity.
The reasons for the different activities could be clearly ascribed to the different nature
of the substituents. Size, conformational and electronic differences explained the
different behaviors. Looking at the AMP derivatives activities, not complicated by the
presence of NNS activity, it was evident that iodine imparted to 7a a certain affinity for
the enzyme active site, hence its ability to act as an inhibitor; in contrast, the phenyl
group in 7b did not generate any inhibition, indicating a low affinity/binding to the
enzyme. These effects are likely to be due to the different size of the substituents and,
therefore, to their ability to fit into the hydrophobic tunnel of the enzyme in
correspondence to the C-2 of the adenine ring. Indeed, iodine and phenyl groups have a
substantially different van der Waals volume (Vw), respectively 19.64 and 45.84
cm3·mol
-1;264
therefore, they would have a different occupancy in the hydrophobic
tunnel of NAD+-dependent DNA ligases. The phenyl substituent, larger than iodine,
Chapter 5
161
could not fit properly into the tunnel, disfavouring the stabilization of 7b in the active
site; this would justify its low binding to EcLigA and, therefore, its absent inhibitor
activity. Instead, iodine could fit more easily into the tunnel, causing the higher affinity
of 7a for EcLigA and, therefore, its activity as inhibitor.
The conformation of iodine and phenyl substituted adenine related to the ribose has to
be considered when comparing the different activities. In 3.1.1, an adopted anti
conformation in solution for the 2-substituted AMP derivatives was confirmed by 1H
NMR analysis, which tended toward a syn conformation for the 2-substituted NAD+
derivatives (see 3.1.2), in particular for the 2-phenyl NAD+, 13b. In the active site of
NAD+-dependent DNA ligases, the adenine of the natural β-NAD
+ assumes a high syn
conformation (Figure 5.10 a); in fact, in the literature it is known that often
nucleosides/nucleotides with high syn or high anti conformations have high biological
activities.176, 265
Therefore, 2-substituted NAD+ derivatives should be able to assume a
high syn conformation in the active site, where the substituents point far from the ribose
avoiding any steric hindrance (Figure 5.10 b). A more pronounced high syn
conformation for 2-phenyl NAD+, 13b, due to the larger volume occupied by the phenyl
substituent, could be responsible for its higher similarity to the natural β-NAD+, leading
to its high substrate activity.
Figure 5.10 β-NAD+ (a) and 13b (b) in the active site of DNA ligase from E. faecalis (PDB 1TAE). 13b
is built on the β-NAD+ structure. The high syn conformation of the adenine ring related to the ribose
tolerates the introduction of substituents in position 2 without interfering with the binding in the active
site. Visualization by PyMol software.
Chapter 5
162
Examining the activities of 13a and 13b, both iodine and phenyl substituted NAD+
derivatives had NNS activities; this indicated the essential nature of the β-NAD+
scaffold in stabilizing in the active site the 2-phenyl AMP moiety, which alone did not
bind properly to the enzyme. The different turnover of the two NAD+ derivatives used
as NNSs by EcLigA could be explained as due to electronic differences between the
iodine and phenyl substituted adenine rings. Indeed, iodine is more electron-
withdrawing than the phenyl group and, consequently, it can influence the electronic
properties of the closest atoms in the aromatic ring.176, 265
Then, N-1 and N-3 on the 2-
iodo adenine could undergo a reduction of electron density reflecting on their H-
bonding ability and, therefore, on the conformation assumed by the adenine ring related
to the ribose. As mentioned in 5.1, an essential residue in the active site of EcLigA is
Lys 290,254
which is involved in H-bonding with N-1 on the adenine ring during all the
three catalytic steps. Its presence is fundamental for making contacts with the adenine
ring to keep it in the right position in the active site; its modification in mutants makes
the enzyme almost inactive, suggesting its essential nature. In the case of 2-iodo NAD+,
13a, the reduced electron density on N-1 would negatively influence its ability to form
H-bonds with Lys 290, explaining the hypothesized slow turnover of 13a compared to
β-NAD+ and, therefore, its resulting inhibitory activity. Conversely, 2-phenyl NAD
+,
13b, would keep intact the ability of N-1 to interact with Lys 290; indeed, the phenyl
group would increase the electronic density on N-1, making the H-bonding even
stronger than in the natural β-NAD+. This effect would explain the similarity and
competition of 13b with β-NAD+, and the resulting lack of ligation inhibition overall.
5.1.3 Biological testing of 6-substituted AMP and NAD+ derivatives
The activity of 6-substituted AMP, AMP-morpholidate and NAD+ derivatives was
tested toward EcLigA. Firstly, 6-substituted NAD+ derivatives, 14a and 14b, were
tested as NNSs of EcLigA in the absence of natural β-NAD+ at 250 μM, choosing this
high concentration to ensure detection of any eventual NNS activity. Both compounds
showed no activity as NNSs of EcLigA (data not shown).
Next, 6-substituted AMP/AMP-morpholidate/NAD+ derivatives, respectively 8a-b, 11a,
and 14a-b, were tested as inhibitors of EcLigA. Inhibition studies were performed
Chapter 5
163
testing the compounds at 250 μM, 10 fold higher than the natural β-NAD+, with
sampling at 7 and 15 min (Figure 5.11). From inhibition experiments, the compounds
showed no activity as inhibitors of EcLigA, with the exception of the 6-phenyl AMP-
morpholidate, 11a, which showed 60% inhibition at 250 μM at 7 min. At 15 min, the
ligation for the reaction containing 11a increased to the level of the control reaction.
The estimated IC50 value for 11a was 172.7 ± 4.1 μM (Appendix 8, Figure A15).
Figure 5.11 Inhibition of 6-substituted AMP/AMP-morpholidate/NAD+ derivatives at 250 μM.
Conditions: DNA substrate (0.5 μM), EcLigA (0.05 μM), β-NAD+ (26 μM), 6-substituted derivative (250
μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations
are final concentrations. Reactions incubated at 30 °C under shaking and sampled at 7 and 15 min.
Control reaction at the same conditions without any 6-substituted derivative. Analysis of assay run on
denaturing PAGE and gel visualized by fluorescence imaging. Data analysed using ImageJ. Each data
point is the mean (± SD) of a set of experiments run in triplicate; data plotting with Excel.
5.1.3.1 Biological activities of 6-substituted AMP and NAD+ derivatives
From activity tests of 6-substituted AMP/AMP-morpholidate/NAD+ derivatives, it was
clear that the series of 6-substituted adenine derivatives was inactive toward EcLigA as
inhibitors and NNSs, independent of the nature of the introduced substituent. Indeed,
both phenyl and pyrrolyl substituents failed to show any kind of activity.
Looking at the crystal structure of β-NAD+-bound DNA ligase from E. faecalis (PDB
1TAE), there exists a small polar cavity that would be able to accommodate the
exocyclic amino group in position 6 on the adenine ring, but which would probably be
too narrow to accommodate a phenyl or a pyrrolyl substituent. In addition, since there
are no essential residues in the active site that interact with the amino group in position
6, its modification should not cause any effect on the activity of the enzyme. Therefore,
0
20
40
60
80
100
control 8a 14a 11a 8b 14b
% li
gati
on
7 min
15 min
Chapter 5
164
the lack of inhibitory activity of 6-substituted adenine derivatives could be due to their
inability to strongly interact with the enzyme active site, consequently allowing an
easier displacement by the natural β-NAD+. In addition, the lack of NNS activity could
be due to some distortion of the adenine ring in the active site to accommodate the
introduced substituent, which would not fit into the cavity for the exocyclic amino
group.
5.1.4 Biological testing of 8-substituted NAD+ derivatives
NAD+ derivatives substituted in position 8, 15b,c,e,g,h, were tested as NNSs and
inhibitors of EcLigA. For the NNS activity, the compounds were tested at 250 μM in
the absence of β-NAD+; this concentration, 10 fold larger than that used for the assay
with the natural β-NAD+, was chosen to ensure detection of any substrate activity. All
compounds belonging to this series appeared inactive as NNSs, even after long reaction
times. Indeed, in the attempt to assess their substrate activity, pre-incubation studies of
8-substituted NAD+ derivatives with EcLigA were performed at different temperatures
and reaction times starting the reaction with addition of DNA; nevertheless, none of
these experiments showed any ligation activity (data not shown). In addition, when
natural β-NAD+ was added to the unsuccessful reactions, the ligation immediately went
to completion (Figure 5.12), indicating the enzyme retained full activity and, therefore,
confirming its inability to use 8-substituted NAD+ derivatives as NNSs.
Chapter 5
165
Figure 5.12 NNS activity test of 15h. Conditions: DNA substrate (0.125 μM), EcLigA (0.1 μM), 15h (250
μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations
are final concentrations. Negative control at the same conditions without 15h. Reaction incubated at 30
°C under shaking for 1 h; then, addition of β-NAD+ (26 μM), and reaction stopped and analysed by
denaturing PAGE after 2.5 h. Gel visualization by fluorescence imaging, data analysis by ImageJ, and
data plotting with Excel.
Inhibition studies were performed by testing the compounds at 250 μM in the presence
of β-NAD+ at 26 μM. From these experiments, 8-substituted NAD
+ derivatives showed
weak or no activity at all as inhibitors of EcLigA at 250 μM (Figure 5.13); only 15e and
15h inhibited EcLigA around 40% at 250 μM.
Figure 5.13 Inhibition of 8-substituted NAD+ derivatives at 250 μM. Conditions: DNA substrate (0.125
μM), EcLigA (0.1 μM), β-NAD+ (26 μM), 8-substituted NAD
+ derivative (250 μM) in 1 x buffer (30 mM
Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA); all concentrations are final concentrations.
Reactions incubated for 10 min at 30 °C under shaking. Control reaction at the same conditions without
any 8-substituted NAD+ derivative. Analysis of assay run on denaturing PAGE and gel visualized by
fluorescence imaging. Data analysed using ImageJ. Each data point is the mean (± SD) of a set of
experiments run in triplicate; data plotting with Excel.
0
10
20
30
40
50
60
70
80
90
0 50 100 150 200 250
% li
gati
on
time [min]
15h substrate
neg control
b-NAD+
0
20
40
60
80
100
120
% li
gati
on
control 15b 15c 15e 15g 15h
Chapter 5
166
5.1.4.1 Biological activities of 8-substituted NAD+ derivatives
From activity tests of 8-substituted NAD+ derivatives, it was clear that this series was
practically inactive toward EcLigA as inhibitors and NNSs, independent of the nature of
the introduced substituent. To explain the inactivity of 8-substituted NAD+ derivatives,
a deeper investigation of the enzyme active site was performed.
Looking at the β-NAD+-bound DNA ligase from E. faecalis (PDB 1TAE), a clear steric
hindrance between the position 8 on the adenine of β-NAD+ and a close loop delimiting
the active site can be seen (Figure 5.14 a); consequently, the introduction of a
substituent in position 8 on the adenine ring would cause an unavoidable clash with the
loop (Figure 5.14 b).
Figure 5.14 β-NAD+ (a) and 15h (b) in the active site of DNA ligase from E. faecalis (PDB 1TAE). 15h
is built on the β-NAD+ structure. The introduction of substituents in position 8 on the adenine ring would
interfere with the binding in the active site. Visualization by PyMol software.
As discussed in 3.1.2, 8-substituted NAD+ derivatives are characterized by a
pronounced syn conformation of the adenine ring, where the substituent in position 8
points far from the ribose. Therefore, 8-substituted NAD+ derivatives would already
have the appropriate syn conformation to fit in the active site of DNA ligase with the
same binding mode of β-NAD+; however, the steric clash with the loop would prevent
their initial accommodation into the active site.
This observation raises the question as to whether 8-substituted NAD+ derivatives
would be able to bind in the enzyme active site at all. In an attempt to answer this
Chapter 5
167
question, the docking of a representative compound from the series of 8-substituted
NAD+ derivatives, 15h, into the crystal structure of DNA ligase from E.faecalis (PDB
1TAE) was performed and the resulting structures with the best scores were compared
with the structure of the natural β-NAD+. The best docked structures showed 15h
respectively in anti (a), syn (b) and high syn (c) conformation (Figure 5.15).
Figure 5.15 Docking of 15h (pink) and superimposition with β-NAD+ (green) in the active site of DNA
ligase from E. faecalis (PDB 1TAE). a) Docked 15h in anti conformation; b) docked 15h in syn
conformation; c) docked 15h in high syn conformation. Visualization by PyMol software.
In an attempt to avoid the steric clash between the substituent in position 8 and the loop,
the docking software flipped the adenine of 180° around the N-glycosidic bond, keeping
the ribose fixed in case a) to obtain an anti conformation; or flipped the adenine of 180°
around the N-glycosidic bond and also rotated the ribose in case b) to allow the adenine
to remain in the favoured syn conformation; or in case c), the adenine kept the high syn
conformation without any torsion around the N-glycosidic bond moving back compared
to its position in β-NAD+ to avoid the steric clash of the substituent with the loop.
The assumption of an anti conformation by 15h in case a) would require a large
conformational change and distortion of the nucleobase related to the ribose,
representing an unrealistic situation. Instead, the maintenance of a syn conformation in
case b) owing to the ribose rotation around the N-glycosidic bond is more likely to
occur because of the ribose flexibility, as well as the maintenance of high syn
conformation in case c), due to the backwards movement of the adenosine moiety to
avoid the steric clash with the loop.
Chapter 5
168
Another interesting observation among the different docked structures was that, by
flipping the adenine of 180° around the N-glycosidic bond, C-8 and its substituent
pointed into the hydrophobic tunnel, instead of C-2, as is the case of the natural β-
NAD+. Therefore, both the potential anti and syn (Figure 5.15 a and b respectively)
conformations of the docked 15h would accommodate the pyrrolyl substituent inside
the hydrophobic tunnel, stabilizing their position in the active site.
These reasonable structures assumed by 15h in the active site of DNA ligase from E.
faecalis show the possibility of the compound fitting in the active site. However, in all
these potential structures the contacts with the essential catalytic residues in the binding
site (Table 5.1) are destroyed. This effect could be fatal in terms of any possible binding
and consequent activity as inhibitors and/or NNSs of 8-substituted NAD+ derivatives on
the DNA ligases.
5.1.5 Summary
Summarising all the results obtained by the screening of AMP/AMP-
morpholidate/NAD+ derivatives substituted on C-2, C-6 or C-8 of the adenine ring, it
was possible to conclude that only the series of the 2-substituted adenine derivatives
showed biological activity toward EcLigA as inhibitors and/or NNSs, depending on the
introduced substituent and its interactions within the active site. Indeed, both the
presence of the hydrophobic tunnel in the enzyme structure in which the substituent in
position 2 can be accommodated, and the high syn conformation of the adenine base
related to the ribose, resembling the natural β-NAD+, allowed the stabilization of the 2-
substituted adenine derivatives in the active site and, therefore, their ability to interfere
with the enzyme activity. In contrast, compounds belonging to the series of adenine
derivatives substituted on C-6 and C-8 could not actively interact within the active site,
consequently showing no activity.
To create new antibacterial drugs acting on the activity of NAD+-dependent DNA
ligases, the appropriate choice of substituents introduced in position 2 of the adenine
ring has to be considered. As shown previously in the literature, the introduction of long
alkyl chains that fit into the hydrophobic tunnel blocking the compounds in the active
Chapter 5
169
site allows the development of efficient inhibitors of NAD+-dependent DNA ligases.
73,
74 From our studies, it emerged that the insertion of small electron-withdrawing groups
in position 2 of the adenine ring, that are able to substantially change the H-bonding
ability of the adenine ring could be important for developing good inhibitors. Moreover,
the size of the substituents would be crucial in generating inhibitory activity; indeed,
from our analysis it was suggested that a phenyl group was already too large to properly
fit into the hydrophobic tunnel. This could be a limit to the application of the Suzuki-
Miyaura cross-coupling in modifying the C-2 position on the adenine ring, since only
aryl groups smaller than the phenyl could be possibly inserted in the adenine ring to
generate inhibitors. In addition, as expected because of their inactivity as NNSs, AMP
derivatives would be more suitable backbones than NAD+ derivatives as inhibitors,
since no side substrate activity could interfere in their influence on the enzyme activity.
AMP-morpholidates, 10b and 11a, showed inhibitor activities toward EcLigA
regardless of the introduced substitution and its position on the adenine ring. At the
outset of this project, it was expected that AMP-morpholidates would either act as
NNSs or inhibit the enzyme by covalent modification. Indeed, their phosphate, activated
compared to that in the AMP derivatives, could undergo the nucleophilic attack by the
lysine residue into the active site to form the adenylated-enzyme; then, the latter could
be active and evolve in the next ligation steps, or be inactive and block the ligation.
However, AMP-morpholidates did not show any NNS activity. Moreover, in the
inhibition experiments the ligation increased at later times in the reaction with both 10b
and 11a (Figure 5.8, Figure 5.11); therefore, the inhibition could not be due to covalent
modification of the enzyme, but it is more likely to be due to reversible binding of the
derivatives in the active site. The fact that 2-phenyl AMP-morpholidate, 10b, was active
as an inhibitor in contrast to 2-phenyl AMP, 7b, and that 6-phenyl AMP-morpholidate,
11a, was active as an inhibitor in contrast to all other derivatives belonging to its series,
showed that the modification on the phosphate, and not those on the adenine ring, was
responsible for the inhibition by AMP-morpholidates.
Looking at the perspective for generating biochemical tools for biochemical
investigations, the development of 2-substituted NAD+ derivatives as NNSs for NAD
+-
dependent DNA ligases has been a useful achievement in this project. Indeed, the
Chapter 5
170
different activities of the differently substituted NAD+ derivatives can be related to
structural and mechanistic details of the enzyme, providing a deeper understanding and,
therefore, a more strategic development of inhibitors.
5.2 Sirtuins
Protein acetylation/deacetylation is a fundamental process to regulate protein activity,
localization, interaction, stability and degradation. Consequently, it contributes to
cellular regulation of processes such as transcription, proliferation, apoptosis and
differentiation, occurring in response to signalling pathways. As is the case for the
phosphorylation performed by kinases with a broad substrate specificity,
acetylation/deacetylation seems to involve many biological targets in answer to
particular intracellular and extracellular signals.266
The main and best known targets for acetylation modification are the histones, proteins
found in all eukaryotic cells, involved in the packaging of DNA into chromatin and,
therefore, responsible for gene regulation. Histones contain many basic lysine residues
that can undergo acetylation/deacetylation modification by acetylases/deacetylases.
When histones are acetylated on relevant lysine residues, the resulting charge on these
amino acids becomes neutral overall and, therefore, histones do not interact so well with
the negatively charged DNA; the latter can then be more easily unwound and
transcribed or replicated by RNA or DNA polymerases. In contrast, when histones are
deacetylated on relevant lysine residues, the resulting positive charge on these amino
acids allows them to interact with the negatively charged DNA, winding it more tightly
and avoiding its transcription or replication. The acetylation/deacetylation process is
therefore part of the epigenetic changes in gene expression that occur without modifying
the DNA sequence; thus, acetylases/deacetylases can regulate gene transcription and
modulate gene silencing and transcriptional repression.
Acetylases/deacetylases are better known as histone acetyltransferases (HATs) and
histone deacetylases (HDACs), despite acting on a variety of substrates. HDACs, first
isolated in yeast from Saccharomyces cerevisiae, are divided into four classes: I, II and
IV classes formed by zinc-dependent deacetylases, while III class by NAD+-dependent
Chapter 5
171
deacetylases. The latter class is responsible for gene silencing, from which its name, Sir,
stands for “silent information regulator”.267
In humans, seven NAD+-dependent HDACs have been isolated and, because of their
homology with the yeast Sir family, these are called sirtuins. An increased interest in
sirtuins has developed in the past years because of their involvement in a variety of
biological processes. The transcriptional gene regulation and the implication of sirtuins
in DNA repair are the main processes in which they are involved, together with cellular
stress resistance, therefore ageing control and life span increase; metabolic regulation,
as they influence glucose, fat and insulin metabolism; and apoptosis regulation, thus
their possible implication as anticancer targets. All these processes have been connected
since the discovery that enzymes, such as sirtuins, appear to be involved in the
regulation of them all.268-270
Sir2 from Saccharomyces cerevisiae, belonging to class III of NAD+-dependent
HDACs, has been structurally characterized. The protein consists of a large domain
containing a Rossmann-fold structure, with two small domains inserted; one of them is
a zinc-binding module, not involved in the catalytic cycle (Figure 5.16 a). The β-NAD+-
binding pocket is characterized by three regions, A for the adenosine binding, B for the
nicotinamide ribonucleoside binding and C, a deep region in the catalytic pocket
involved in the nicotinamide cleavage (Figure 5.16 b). When β-NAD+ binds to the
enzyme, NaM would fit in the B region but, in the presence of acetyl lysine, the enzyme
would undergo a conformational change bringing NaM into the C region, where the N-
glycosidic bond is cleaved.
As mentioned in 1.2.1, it is not certain if sirtuins cleave the nicotinamide N-glycosidic
bond by an SN1 mechanism, via formation of the oxocarbenium ion intermediate, or by
an SN2 mechanism, with nucleophilic attack by the oxygen of the acetyl lysine on C-1′′
and simultaneous release of NaM. However, the reaction products are the deacetylated
lysine and the 2′-OAc-ADPR, in equilibrium with the 3′-OAc-ADPR.31, 267, 269
As described in Chapter 1, inhibition or activation of sirtuins activity can regulate
different biological processes and provide understanding of their mode of action. NaM,
one of the sirtuins reaction products, is a non-competitive inhibitor for sirtuins. It seems
Chapter 5
172
to bind in the C region of the catalytic pocket, more than in the B one, causing the
regeneration of β-NAD+ by transglycosidation.
31 Consequently, many sirtuins inhibitors
based on mimicking NaM have been developed, together with others mimicking the
adenosine72
and NAD+ structure (see 1.3.1).
31, 70, 267
Figure 5.16 a) Structure of archaeal Sir2-Af2 with β-NAD+ bound; b) β-NAD
+-binding pocket (PDB
1YC2). Visualization by PyMol software.
Since several biological processes, such as life span increase, seem to be linked to an
activation of the sirtuins activity, the discovery of sirtuins activators has become more
important. A class of polyphenols, resveratrol being the most known and representative,
has shown activator activity toward sirtuins; this effect seems to be connected with the
binding of activators in the C region, where otherwise free NaM released from the
reaction would bind and inhibit the sirtuins activity.271
5.2.1 Biological testing of 8-substituted NAD+ derivatives
As mentioned in Chapter 1, NAD+ derivatives can be effective inhibitors or NNSs of
sirtuins depending on the substitutions performed to modify the structure of the natural
substrate.31
It has been shown that carba-NAD+ (Figure 1.15), where the β-D-ribose of
the nicotinamide ribonucleoside has been replaced with a 2,3-dihydroxycyclopentane
Chapter 5
173
ring, is not a hydrolysable NAD+ derivative and, therefore, acts as an inhibitor of
sirtuins. This compound has been a useful biochemical tool to investigate and
understand the structure of the active site and the sirtuins reaction mechanism.31
8-aryl/heteroaryl NAD+ derivatives, 15b, 15h
246, 16-22, synthesized in Chapter 2 (see
2.7.2, Scheme 2.20 and Figure 5.17), were tested as inhibitors and/or NNSs of human
SIRT1-270
in collaboration with Prof. M. Jung and co-workers. The assay described in
Chapter 1 (see 1.4.7), using Z-MAL as a non-natural peptide substrate for sirtuins, was
used to evaluate the activities of the tested compounds.
Figure 5.17 8-aryl/heteroaryl NAD+ derivatives, 15b, 15h, 16-22, tested as inhibitors and/or NNSs of
human SIRT1-2.70
From the inhibition experiments with 8-aryl/heteroaryl NAD+ derivatives and a few of
the corresponding adenosine and AMP derivatives at 50 μM, it emerged that, with the
exception of 15h246
and 20, all 8-aryl/heteroaryl NAD+ derivatives exerted
preferentially an inhibitory activity toward SIRT2 compared to SIRT1. For those
compounds showing an inhibitory activity ≥ 50% at 50 μM on either sirtuins isoforms,
IC50 values were estimated and found to be in the order of low micromolar
concentration.70
In addition to the 8-aryl/heteroaryl NAD+ derivatives, the inhibitory
activity was tested for the corresponding adenosine and AMP derivatives with bromo
and phenyl substituents. With the exception of 8-phenyl AMP, all other derivatives
showed no activity toward either SIRT1 nor SIRT2, indicating the necessity to have the
Chapter 5
174
whole β-NAD+ scaffold to be effective inhibitors.
70 With the hypothesis that 8-
aryl/heteroaryl NAD+ derivatives could act as NNSs for sirtuins and, consequently,
influence the inhibition results, their substrate activity was also tested. With the
exception of 15b, which showed a minor substrate activity, all other derivatives did not
have any.70
5.2.2 Biological activities of 8-substituted NAD+ derivatives
Investigations using computational studies were undertaken to provide a better
understanding of the mode of inhibition of 8-aryl/heteroaryl NAD+ derivatives and their
apparent selectivity toward the different isoforms of human sirtuins. The alignment of
sequences for SIRT1 and SIRT2 was used to analyse the similarities between the
different isoforms; effectively, only a 44% identity was observed between the two
enzymes.70
Examining the crystal structures available for human SIRT2 (PDB 1J8F) and SIRT3
(PDB 3GLT), a flexible loop with a high level of sequence conservation has been
reported to be involved in the β-NAD+ binding and NaM release.
31 However, in the
different isoforms the loop seems to assume completely different conformations.
Therefore, interferences between the substituents in position 8 and the flexible loop
could be responsible for the selectivity of 8-aryl/heteroaryl NAD+ derivatives toward
the different isoforms of human sirtuins and for their lack of substrate activity.70
The few available crystal structures for different sirtuins isoforms containing β-NAD+
or ADPR show an adenine in anti conformation compared to the ribose. However, as
discussed in 3.1.2, all 8-aryl/heteroaryl NAD+ derivatives have the adenine ring in a
more or less definite syn conformation because of the steric hindrance between the
substituent in position 8 and the ribose.70
From the docking of a representative 8-
substituted NAD+ derivative (17 in Figure 5.17) within SIRT2 (PDB 1J8F) and
superimposition with β-NAD+ in the archaeal enzyme Sir2-Af2 (PDB 1YC2), the same
anti conformation was observed for adenine in 17 and β-NAD+, suggesting that 8-
aryl/heteroaryl NAD+ derivatives could undergo conformational changes during their
binding to the enzyme. Sirtuins themselves could be able to induce these conformational
Chapter 5
175
changes with a different ability depending on the isoform involved, consequently
influencing the inhibition selectivity of 8-aryl/heteroaryl NAD+ derivatives toward the
different sirtuins isoforms. In support of this hypothesis, it was observed that 15h246
and
20, characterized by a strong syn conformation (Table 3.2), had no inhibitor and
substrate activities,70
presumably because sirtuins would not be able to overcome such
definite conformations of the molecule.
Chapter 6
176
6 CONCLUSIONS AND FUTURE WORK
NAD+-consuming enzymes play a crucial role in important biological processes; hence,
it is fundamental to require a better understanding of their biological activities, how
they are modulated and how they influence each other and the activities of other
important enzymes in cells. The development of NAD+ derivatives as biochemical tools
for the study of this class of enzymes has been, and is still, one of the most challenging
biochemical topics.
Reviewing the substitutions performed on the existing NAD+ derivatives and their
influences on the enzyme activities, the modification of the adenine ring appeared as the
most suitable to confer selective biological activities as inhibitors and/or NNSs to the
novel NAD+ derivatives toward a selected range of enzymes. Due to the recent and
impressive advances in the application of palladium-mediated cross-couplings to the
nucleosides/nucleotides modifications, the choice of using the Suzuki-Miyaura cross-
coupling to substitute the adenine ring made possible the development of aryl/heteroaryl
adenine-modified NAD+ derivatives. In particular, positions 2, 6 and 8 were modified
with phenyl, and, where possible, pyrrolyl groups to explore the different properties of
the compounds and their activities generated by the introduction of hydrophobic and H-
bond making groups.
The successful synthesis of aryl/heteroaryl NAD+ derivatives substituted on C-2, C-6 or
C-8 of the adenine ring was accomplished following the optimization of the known
multistep synthesis to generate the new classes of compounds. However, two additional
strategies, requiring further studies, were investigated to accelerate the synthesis of the
NAD+ derivatives. The pathway via phosphoramidite nucleoside intermediate was
unsuccessful presumably because of the low nucleophilicity and insolubility of β-NMN
in acetonitrile. Therefore, beyond a screening of polar aprotic solvents (e.g. DMSO,
DMF, acetone) to improve the solubility of β-NMN, the replacement of DCI with other
phosphoramidite activators (e.g. tetrazole, HOBt) during the coupling with β-NMN
could bring the reaction to success. On the other hand, the pathway via β-NAD+
bromination/Suzuki-Miyaura cross-coupling in a one-pot, two-step procedure allowed
Chapter 6
177
the successfull synthesis of different 8-aryl/heteroaryl NAD+ derivatives, although it did
fail in the coupling with very electron poor boronic acids.
The synthesized compounds were analysed from the spectroscopic and conformational
perspective, showing interesting properties. In particular, 8-aryl/heteroaryl adenine
derivatives showed fluorescent properties that could be modulated depending on the
nature of the substituent introduced. In addition, as shown from our studies and previous
data about the conformational properties of β-NAD+ and its well known derivative, ε-
NAD+, all of the synthesized NAD
+ derivatives were characterized by a folded
conformation in aqueous solution, due to the interactions between the hydrophobic
aromatic rings in the molecule. This folded conformation could be released by solvation
in organic solvents, bringing the opening of the β-NAD+ scaffold to an extended
conformation. These conformational properties, combined with the fluorescence of 8-
aryl/heteroaryl NAD+ derivatives, were the basis of their use as fluorescent probes to
monitor the activities of NAD+-consuming enzymes. Indeed, the folded conformation of
the compounds caused a fluorescence quenching that could be disrupted by the
separation of the modified adenine ring, the fluorophore under discussion, and the
nicotinamide moiety with consequent fluorescence re-establishment. The enzymatic
cleavage of the β-NAD+ scaffold on the pyrophosphate or the nicotinamide N-glycosidic
bonds, catalyzed by NAD+-consuming enzymes, was therefore responsible for a
fluorescence increase allowing monitoring of the enzymatic reactions.
The use of fluorescent 8-substituted NAD+ derivatives as NNSs by NAD
+-consuming
enzymes made them useful biochemical probes to investigate enzymatic structures,
activities and kinetics in more detail by monitoring their fluorescence changes.
Therefore, fluorescence-based assays were successfully developed for selected NAD+-
consuming enzymes, giving interesting mechanistic information.
The fluorescence principle of the assays using 8-aryl/heteroaryl NAD+ derivatives is
essentially the same as those developed with the well known ε-NAD+. However, the use
of our best fluorophore, 15h, containing a pyrrolyl group in position 8 on the adenine
ring, increased the sensitivity of the assays compared to those using ε-NAD+ because of
a stronger fluorescence increase during the enzymatic reaction. Moreover, the
Chapter 6
178
application of 15h for the monitoring of enzymes for which ε-NAD+ failed in the
development of continuous fluorescence-based assays could be investigated. For
instance, as discussed in Chapter 1, ε-NAD+ was not a good substrate for PARPs and a
continuous fluorescence-based assay could not be developed. Therefore, it would be
worth testing 15h on PARPs activity to assess whether there were positive differences
compared to ε-NAD+.
In contrast to the 8-aryl/heteroaryl NAD+ derivatives, those substituted in position 2 and
6 showed weak, or no fluorescence at all; therefore, they could not be used as
fluorescent biochemical probes. Nevertheless, they were still useful tools to investigate
enzymatic structures and mechanisms. In particular, the biological activities of 2-
substituted NAD+ derivatives toward DNA ligases gave clues to fundamental
interactions between the β-NAD+ scaffold and its binding site.
It was shown that 2-iodo and 2-phenyl NAD+ derivatives, both non-fluorescent, were
NNSs of EcLigA. Therefore, an idea for developing fluorescent NNSs came from the
considerations of their activities. Since a substituent in position 2 on the adenine ring is
crucial to confer to the NAD+ derivatives the ability to bind in the active site by
positioning the substituent into the hydrophobic tunnel, as well as to assume a high syn
conformation resembling that of the natural b-NAD+, the substitution of the adenine
ring with 2-amino purine (1) or 2,6-diamino purine (2)190
(Figure 6.1), both isomorphic
fluorescent nucleobases (Chapter 3, see 3.3.2), may confer to the corresponding NAD+
derivatives their fluorescence properties.
Figure 6.1 Two possible fluorescent NNSs of NAD+-dependent DNA ligases: nicotinamide 2-amino
purine dinucleotide (1) and nicotinamide 2,6-diamino purine dinucleotide (2).
Chapter 6
179
These compounds could potentially be exploited for binding and kinetic/mechanistic
studies. The binding studies could rely on the fluorescence enhancement of the NNSs
following their passage from a folded quenched conformation in solution to an extended
conformation in the active site. On the other hand, the DNA ligase activity could be
monitored by the fluorescence increase due to the formation of the AMP derivative
from the cleavage of the quenched NAD+ derivative (the same principle on which the
monitoring of NPPs is based). In addition, kinetic and mechanistic details could come
from the fluorescence changes occurring in the fluorescent adenine derivatives
following their interactions with Lys 290 in the active site along the three different
reaction steps. Indeed, the formation of the H-bond between N-1 and Lys 290 would
change the electronic delocalization along the adenine-modified rings; hence, the
formation of a partial positive charge on N-1 could bring a fluorescence quenching
similar to that occurring in the N1-cADPR derivatives analysed in Chapter 4.
Of course, these modifications could have some limitations. Firstly, the introduced
amino group in position 2 is polar, so it is unlikely to properly interact with the
hydrophobic tunnel of the DNA ligase, and could be unable to stabilize the NAD+
derivatives in the active site. Secondly, 2-AP would lack the exocyclic amino group of
the adenine ring, which could cause a decreased binding in the active site for this
compound. However, since no fluorescent NAD+ derivatives have been reported as
NNSs for NAD+-dependent DNA ligases (ε-NAD
+ and 8-aryl/heteroaryl NAD
+
derivatives did not show any activity), it would be worth exploring the activity of these
two compounds in order to create fluorescent probes for NAD+-dependent DNA ligases.
A comparison of activities exerted by 8-aryl/heteroaryl NAD+ derivatives against all of
the enzymes examined in this project was undertaken to investigate their selectivity and,
therefore, their potential application for in vitro/in vivo studies. A summary of the
biological evaluation of the 8-substituted NAD+ derivatives against the different NAD
+-
consuming enzymes is shown in Table 6.1.
Chapter 6
180
Activity
Enzyme Inhibitor Substrate
NPP (Crotalus adamanteus venom)
NADase (porcine brain)
ADPRC (Aplysia californica)
Human SIRT1-2 –
DNA ligase (EcLigA) – –
Table 6.1 Activities of 8-aryl/heteroaryl NAD+ derivatives toward different NAD
+-consuming enzymes.
The tick indicates that the compounds showed the specified activity toward the selected enzyme; the
line – indicates that the compounds did not show the specified activity toward the selected enzyme.
A certain selectivity exerted by 8-aryl/heteroaryl NAD+ derivatives against the different
enzymes was observed. Indeed, with the exception of NADase and ADPRC sharing the
same unifying catalytic mechanism (Chapter 4, see 4.2), the enzymes examined, though
sharing the same substrate/co-substrate, have different active sites and modes of β-
NAD+ binding, which are responsible for their different catalytic reactions and,
therefore, for the different activities shown by 8-substituted NAD+ derivatives. To
underline the intrinsic structural differences that distinguish these enzymes and their
activities, the available crystal structures for β-NAD+-bound ADPRC from Aplysia
californica (a), archaeal Sir2-Af2 (b) and DNA ligase from E. faecalis (c) were
examined and their active sites were compared (Figure 6.2).
Figure 6.2 β-NAD+ in the active site of ADPRC from Aplysia californica (PDB 3ZWM) (a), archaeal
Sir2-Af2 (PDB 1YC2) (b), DNA ligase from E. faecalis (PDB 1TAE) (c). Visualization by PyMol
software.
In ADPRC (Figure 6.2 a), the β-NAD+ binding site is completely exposed to the solvent
and β-NAD+ is bound in a folded conformation, since the enzyme catalyzes its
cyclization to cADPR. The adenine ring, free from any interaction with the enzyme and
originally in an anti conformation related to the ribose, during the catalytic cycle will
Chapter 6
181
orient in a syn conformation to facilitate the cyclization on N-1 (Table 6.2). Due to the
lack of interactions of the adenine ring with the enzyme, the substitution in position 8 in
the synthesized NAD+ derivatives could be tolerated by the enzyme and the compounds
could fit in the active site already in a syn conformation, which facilitates the
cyclization.
In the archaeal Sir2-Af2 (Figure 6.2 b), the β-NAD+ binding site is quite exposed to the
solvent and β-NAD+ is bound in an extended conformation. The adenine ring, in anti
conformation related to the ribose (Table 6.2), interacts strongly with the active site
residues though not directly involved in the catalytic reaction. Therefore, its
modification interferes with the enzyme activity, making 8-substituted NAD+
derivatives inhibitors of sirtuins.
Finally, in DNA ligase (Figure 6.2 c), the β-NAD+ binding site is exposed to the solvent
at the moment of the binding and β-NAD+ is bound in an extended conformation. The
adenine ring, directly involved in the catalytic reaction (see 5.1), interacts with the
active site residues. Its initial high syn conformation in β-NAD+ will turn into anti in the
adenylated-enzyme intermediate (1) (Scheme 5.1), to finally return to high syn in the
adenylated-DNA intermediate (2) (Scheme 5.1, Table 6.2). Hence, in the DNA ligase
catalytic mechanism the adenine moiety is key for the enzymatic reaction and the β-
NAD+ binding is so specific that substitutions on the adenine ring, unless the adenine
orientation in β-NAD+ is left unchanged, do not allow any enzymatic activity. Indeed,
NAD+ derivatives substituted on C-6 and C-8 of the adenine ring, in which the
substituents change the position/orientation of the adenine in the active site, did not
exert any effect on the activity of EcLigA because they could not bind in any active way
to the enzyme.
Just from this simple visual analysis of the active site and β-NAD+ binding mode, it is
possible to suggest a structural and, therefore, biological diversity between the three
examined enzymes, justifying the different biological activity exerted by 8-substituted
NAD+ derivatives.
Chapter 6
182
Enzyme Adenine conformation changes during the catalytic cycle
ADPRC
(Aplysia
californica)
Human
SIRT1-2
DNA ligase
(EcLigA)
Table 6.2 Changes in the adenine conformation during the catalytic cycle in ADPRC, SIRT1-2 and DNA
ligase.
For the examined structures of ADPRC from Aplysia californica, archaeal Sir2-Af2 and
DNA ligase from E. faecalis, a sequence alignment was performed, showing only 0.9%
of identity. Therefore, structure similarities were estimated pairwise by comparing the
results obtained with rigid and flexible structure alignments (Appendix 8, Table A1).272,
273 The latter method allows consideration of proteins as flexible bodies, as they are in
reality; hence, the program, instead of just answering how large is the similar part that
two proteins share, as happens in the rigid structure alignment, it also estimates how
much a protein can rearrange itself to be more similar to another one. All these analyses
were in agreement with a deep structural diversity of the enzymes examined, confirming
once again their different biological roles, though sharing the same substrate/co-
Chapter 6
183
substrate, β-NAD+. This structural diversity definitively reflected on the various
activities exerted by 8-substituted NAD+ derivatives toward these enzymes.
Summarising the results obtained in this project, it is possible to conclude that
aryl/heteroaryl substitutions on different positions on the adenine ring can generate
novel NAD+ derivatives with biological activities toward NAD
+-consuming enzymes.
The preferential action of aryl/heteroaryl adenine-modified NAD+ derivatives as
inhibitors and/or NNSs toward specific NAD+-consuming enzymes offers the
prospective of their use for in vitro/in vivo studies. Of course, such a step forward in
their biological evaluation means considering many factors that until now have not been
an issue. For instance, in the case of their activity on intracellular enzymes, such as
sirtuins or DNA ligases, their resistance to the degradation exerted by ecto NAD+-
consuming enzymes or intracellular nucleases should be considered. Therefore,
chemical modifications on the β-NAD+ scaffold or alternative delivering methods
should be developed related to the use for which they are designed. However, the
validity of their use as biochemical probes is already a great pace to widen the
knowledge of fundamental biological processes involving NAD+-consuming enzymes.
Chapter 7
184
7 EXPERIMENTAL
7.1 General
7.1.1 Synthesis and characterization
All synthetic reagents were obtained commercially and used as received, including
anhydrous solvents over molecular sieves, unless otherwise stated. Milli-Q water was
used for the preparation of aqueous solutions and as solvent in the reactions. For the
synthesis of phosphoramidites, CH2Cl2 was distilled over calcium hydride, and DIPEA
was dried over molecular sieves (4 Å) for four days before the use. For the preparation
of dinucleotides, MnCl2 solution in formamide was dried over molecular sieves (4 Å)
and stored under nitrogen atmosphere. All reagents and solvents used for analytical
applications were of analytical quality. TLC was performed on precoated slides of Silica
Gel 60 F254 (Merck). Spots were visualized under UV light (254/365 nm). Reaction
products were characterized by high-resolution mass spectrometry (HR-MS), 1H,
13C
and 31
P NMR spectroscopy, and HPLC. 1H NMR spectra were recorded at 298 K on a
Varian VXR 400 S spectrometer at 400 MHz, or on a Bruker Avance DPX-300
spectrometer at 300 MHz, or on a Bruker BioSpin GmbH at 400 MHz. 13
C NMR
spectra were recorded at 300 K on a Bruker BioSpin GmbH at 150.90 MHz and at 101
MHz. 31
P NMR spectra were recorded at 298 K on a Varian VXR 400 S spectrometer at
161.98 MHz, or on a Bruker BioSpin GmbH at 161.98 MHz. Chemical shifts (δ) are
reported in ppm (parts per million) and referenced to D2O, CD3OD or CDCl3
(respectively, δH 4.79, 3.31, 7.26) for 1H NMR, and to CH3 of TEA, CD3OD or
CDCl3 (respectively, δC 9.07, 49.00, 77.16) for 13
C NMR. Coupling constants (J) are
reported in Hz. Accurate electrospray ionization mass spectra were obtained on a
Finnigan MAT 900 XLT mass spectrometer at the EPSRC National Mass Spectrometry
Service Centre, Swansea. Ion-pair and ion-exchange chromatography was performed on
a Biologic LP chromatography system equipped with a peristaltic pump and a 254 nm
UV Optics Module. For detailed conditions, see 7.1.2. High performance liquid
chromatography (HPLC) was performed on a Perkin Elmer series 200 equipped with a
Supelcosil LC-18T column (25 cm x 4.6 mm, particle size 5 μm), a diode array detector
(detection wavelengths: 254/280 nm) and a column oven (temperature: 30 ºC). For
Chapter 7
185
detailed conditions, see 7.1.2. Absorbance spectra were recorded on a PerkinElmer
Lambda 25 UV/Vis spectrometer at room temperature in FarUV quartz cells (path
length 1.0 cm). Fluorescence spectra were recorded on a PerkinElmer LS-45
spectrometer at room temperature in a quartz fluorescence cell (path length 1.0 cm).
7.1.2 Chromatography conditions
Preparative chromatography – purification method 1: Ion-pair chromatography was
performed using Lichroprep RP-18 resin and a variable gradient of methanol against
0.05 M TEAB (triethylammonium bicarbonate buffer). Product containing fractions
were combined and reduced to dryness. The residue was co-evaporated repeatedly with
methanol to remove residual TEAB.
Preparative chromatography – purification method 2: Anion-exchange chromatography
was performed using BioscaleTM
Mini Macro-Prep High Q cartridges and a gradient of
0-100% 1 M TEAB (pH 7.3) against Milli-Q H2O. Product containing fractions were
combined and reduced to dryness. The residue was co-evaporated repeatedly with
methanol to remove residual TEAB.
High performance liquid chromatography (HPLC): samples (injection volume: 20 μL)
were analysed using a mobile phase composed of 0.05 M triethylammonium phosphate
buffer at pH 8 (solvent A) and methanol (solvent B), according to the gradient steps in
Table 7.1 (flow rate: 1.5 mL/min).
Step Minutes (min) Gradient
1 0 – 2 90% A – 10% B
2 2 – 17 10 – 50% B
3 17 – 19 50% B
4 19 – 21 50 – 10% B
5 21 – 25 90% A – 10% B
Table 7.1 Gradient steps for HPLC analysis of the NAD+ derivatives.
Chapter 7
186
7.1.3 Enzymology
All reagents used for biological application were of analytical quality and solutions
prepared with Milli-Q water. NPP from Crotalus adamanteus venom, NADase from
porcine brain and ADPRC from Aplysia californica were purchased from Sigma.
Human SIRT1-2 were expressed and purified as described70, 274
with minor
modifications. EcLigA from E. coli was expressed and purified as described250
with
minor modifications (see 7.7.1). pET-16b vector, His-tag binding column (IDA-based
His-Bind Resin Ni2+
precharged) and its buffer kit were purchased from Novagen.
Protein desalting was performed using a PD-10 Column (Sephadex G-25 Medium) and
its buffer (20 mM Tris, 200 mM NaCl, pH 7.5). To quantify the protein concentration
according to the Bradford method, the Bio-Rad Protein Assay, consisting of a protein
dye reagent (Coomassie Brilliant Blue G-250 in acidic solution), was purchased from
Bio-Rad (absorbance reading at 595 nm). The oligonucleotides for the ligation assay
were supplied by Eurogentec. For fluorescence imaging, Molecular Dynamics Storm
Phosphorimager was used. Fluorescence assays were performed in NUNC 96 plates on
a BMG labtech PolarStar microplate reader equipped with suitable absorbance and
emission filters.
7.2 Synthesis
7.2.1 General synthetic procedures
A Iodination of C-2 by diazotization-iodine substitution134
Isoamyl nitrite (3.1 equiv.) was added to a mixture of respectively 1a and 1c (1 equiv.)
with I2 (1 equiv.), CH2I2 (10.4 equiv.) and CuI (1 equiv.) in the given volume of THF.
The mixture was refluxed for the given time and then, cooled down at room
temperature. Inorganic salts were removed by filtration, and the filtrate was evaporated
to dryness in vacuo. The crude product was purified by silica gel chromatography, as
indicated for each compound.
Chapter 7
187
B1 Suzuki-Miyaura cross-coupling on C-6158
6-chloro purine nucleoside (5) (1 equiv.), Na2PdCl4 (0.03 equiv.), TPPTS (2.5 equiv. to
Pd), K2CO3 (2 equiv.) and aryl/heteroaryl boronic acid (1.2 equiv.) were placed in a
flask and purged with nitrogen for 15 min. Degassed water was added by syringe to the
reaction vessel and the mixture was stirred at 100 °C under nitrogen for the given time.
When the reaction was complete, the mixture was evaporated in vacuo, and the crude
product purified as indicated for each compound.
B2 Suzuki-Miyaura cross-coupling on C-8172
8-Br-AMP (9a) (1 equiv.) and the appropriate amounts of Na2PdCl4, TPPTS, K2CO3 or
K3PO4, and aryl/heteroaryl boronic acid were placed in a flask and purged with nitrogen
for 15 min. Degassed water was added by syringe to the reaction vessel and the mixture
was stirred at the appropriate temperature under nitrogen for the given time. When the
reaction was complete, the mixture was evaporated in vacuo, and the crude product
purified by chromatography, as indicated for each compound.
C 2′,3′-O-isopropylidene protection141
The respective nucleoside (1 equiv.) was suspended in acetone and 70% HClO4 (2.5
equiv.) was added to the mixture. The reaction was stirred at room temperature for 1.5
h. Then, NH4OH 25% (5.6 equiv.) was added to the reaction, and the solution stirred for
3 h. The mixture was diluted with EtOAc and extracted with brine (3x). Then, the
organic phase was dried over anhydrous MgSO4 and evaporated to dryness in vacuo.
The crude product was purified as indicated for each compound.
D1 Phosphorylation of 5′-OH by POCl3 (method 1)136, 144
The respective 2′,3′-O-isopropylidene protected nucleoside (1 equiv.) was dissolved in
the given solvent with optional proton sponge, and the solution cooled down to 2 °C in
ice-bath. Then, the appropriate amount of POCl3 was added drop-wise to the reaction
mixture, that was kept at 4 °C for 12 h. The reaction was quenched adding drop-wise 1
M TEAB buffer. If proton sponge was present, the mixture was diluted with EtOAc and
Chapter 7
188
extracted (3x). Then, the aqueous solution was evaporated to dryness in vacuo and the
crude product purified by purification method 2, as indicated for each compound.
D2 Phosphorylation of 5′-OH via phosphoramidite (method 2)113
The respective 2′,3′-O-isopropylidene protected nucleoside (1 equiv.) was dissolved in
anhydrous CH2Cl2 and co-evaporated twice to dryness. Then, it was dissolved again in
anhydrous CH2Cl2, and DIPEA (2.5 equiv.) and the given quantity of 2-cyanoethyl N,N-
diisopropyl chlorophosphoramidite were added via syringe. The reaction was stirred
under nitrogen at room temperature for the given time. Then, 5.0 – 6.0 M tBuOOH in
nonane solution (4 equiv.) was added to the reaction and stirred for the given time. The
mixture was diluted with EtOAc and extracted with a saturated sodium bicarbonate
solution (2x) and brine (1x). The organic phase was dried over anhydrous MgSO4 and
evaporated to dryness in vacuo; then, the residue was suspended in NH4OH 25% (30
equiv.) under stirring at room temperature overnight. The solution was evaporated to
dryness in vacuo, and the crude product used for the next step (see procedure E).
E 2′,3′-O-isopropylidene deprotection136
The respective 2′,3′-O-isopropylidene protected nucleotide was dissolved in water and
stirred with Dowex 88 (H+) (ca 0.1 g) overnight. After filtration, the solution was
evaporated to dryness in vacuo and the crude product purified by purification method 1,
as indicated for each compound.
F Synthesis of AMP-morpholidate derivatives112, 172
The respective nucleotide (1 equiv.) was dissolved in DMSO, and then co-evaporated
with DMF (3x). The residue was dissolved in anhydrous DMSO and morpholine (6
equiv.) was added. After 5 min, 2,2′-dithiopyridine (3 equiv.) was added and,
subsequently, triphenylphosphine (3 equiv.). The yellow solution was stirred at room
temperature under nitrogen until TLC showed complete conversion. The product was
isolated by purification method 1, or by precipitation as sodium salt with NaI in acetone
(0.1 M), as indicated for each compound. The NaI solution was added drop-wise to the
reaction mixture until a solid precipitate formed; then, the precipitate was washed
Chapter 7
189
repeatedly with acetone until the solution was colourless. The crude product was dried
and used in the next reaction step without further purification.
G Synthesis of NAD+ derivatives112, 172
The appropriate amounts of AMP-morpholidate and β-NMN were co-evaporated with
pyridine (3x). The flask with the residual solid was purged with nitrogen for 15 min.
Anhydrous MgSO4 (2 equiv.) and MnCl2 (0.2M in formamide, 1.5 equiv.) were added
and the reaction was stirred at room temperature under nitrogen until TLC showed
complete conversion. The reaction was stopped by drop-wise addition of acetonitrile
until a white precipitate formed. The supernatant was removed and the solid dissolved
in water and evaporated to dryness in vacuo. The crude product was purified by
chromatography, as indicated for each compound.
7.2.2 Preparation of 2-substituted AMP derivatives (7a-c)
2-Amino-6-chloro-9β-(2′,3′,5′-tri-O-acetyl-D-ribofuranosyl)-purine (1a)134
Triethylamine (8.7 equiv.), DMAP (12 equiv.) and acetic anhydride (10.7 equiv.) were
added to a suspension of the commercially available 1c (0.4465 g, 1 equiv.) in
acetonitrile (30 mL). The clear mixture was stirred at room temperature for 1 h, then
quenched with MeOH and evaporated to dryness in vacuo. The crude product was
extracted with EtOAc/water (2x) and the organic phase was dried over anhydrous
MgSO4. After concentration to dryness in vacuo, the residue was crystallized from
MeOH to give 1a as white crystals (0.5635 g, 89% yield). 1H NMR (400 MHz; CDCl3)
δ: 7.87 (1H, s, 8-H), 6.00 (1H, d, J1′,2′ = 4.9 Hz, 1′-H), 5.95 (1H, apparent t, 2′-H), 5.74
(1H, apparent t, 3′-H), 4.51 – 4.31 (3H, m, 4′-H, 5′-H2), 2.14 (3H, s, Ac), 2.10 (3H, s,
Chapter 7
190
Ac), 2.09 (3H, s, Ac); 13
C NMR (101 MHz; CDCl3) δ: 170.66, 169.73, 169.50, 164.16,
159.20, 152.09, 140.82, 86.75, 80.14, 72.86, 70.61, 63.06, 51.05, 20.87, 20.70, 20.58.
2-Iodo-6-chloro-9β-(2′,3′,5′-tri-O-acetyl-D-ribofuranosyl)-purine (1b)134
The title compound was prepared according to the general synthetic procedure A, from
1a (0.1257 g, 1 equiv.), isoamyl nitrite, I2, CH2I2 and CuI in THF (1.5 mL). The mixture
was refluxed for 12 h and then, cooled down at room temperature. The crude product
was purified by silica gel chromatography, washing the column with CH2Cl2 to remove
the iodine, then with an eluent 98:2 CH2Cl2/MeOH. 1b was obtained as white crystals
(0.1349 g, 85% yield). 1H NMR (400 MHz; CDCl3) δ: 8.20 (1H, s, 8-H), 6.18 (1H, d,
J1′,2′ = 5.4 Hz, 1′-H), 5.76 (1H, apparent t, 2′-H), 5.57 (1H, m, 3′-H), 4.59 – 4.22 (3H, m,
4′-H, 5′-H2), 2.15 (3H, s, Ac), 2.11 (3H, s, Ac), 2.08 (3H, s, Ac); 13
C NMR (101 MHz;
CDCl3) δ: 170.29, 169.64, 169.48, 152.02, 151.02, 143.28, 132.29, 117.06, 86.72,
80.88, 73.41, 70.61, 62.99, 20.94, 20.64, 20.50.
2-Iodo-6-chloro-9β-D-ribofuranosyl-purine (1d)
The title compound was prepared according to the general synthetic procedure A, from
the commercially available 1c (0.5544 g, 1 equiv.), isoamyl nitrite, I2, CH2I2 and CuI in
THF (40 mL). The mixture was refluxed for 4 h and then, cooled down at room
temperature. The crude product was purified by silica gel chromatography, washing the
Chapter 7
191
column with CH2Cl2 to remove the iodine, then with an eluent 90:10 CH2Cl2/MeOH. 1d
was obtained as a colourless oil (0.382 g, 51% yield). 1H NMR (400 MHz; CD3OD) δ:
8.77 (1H, s, 8-H), 6.09 (1H, d, J = 5.0 Hz, 1′-H), 4.65 (1H, apparent t, 2′-H), 4.36 (1H,
apparent t, 3′-H), 4.15 (1H, m, 4′-H), 3.90 (1H, dd, J5′a,4′ = 3.1 Hz, J5′a,b = 12.3 Hz, 5′a-
H), 3.79 (1H, dd, J5′b,4′ = 3.5 Hz, J5′b,a = 12.3 Hz, 5′b-H). 6-Chloro-9β-D-ribofuranosyl-
purine was eluted later as a side-product of the reaction. 1H NMR (400 MHz; CD3OD)
δ: 8.86 (1H, s, 2-H), 8.75 (1H, s, 8-H), 6.16 (1H, d, J = 5.2 Hz, 1′-H), 4.71 (1H,
apparent t, 2′-H), 4.38 (1H, apparent t, 3′-H), 4.16 (1H, m, 4′-H), 3.90 (1H, dd, J5′a,4′ =
3.0 Hz, J5′a,b = 12.3 Hz, 5′a-H), 3.79 (1H, dd, J5′b,4′ = 3.3 Hz, J5′a,b = 12.3 Hz, 5′b-H).
2-Iodo-9β-D-ribofuranosyl-adenine (4)134
1b (0.1304 g, 1 equiv.) and 1d (0.382 g, 1 equiv.) were respectively dissolved in 7 N
methanolic ammonia and heated at 60 °C for 12 h in a sealed tube. After degassing the
solution, the latter was evaporated to dryness in vacuo; then, the residue was
crystallized from water to give 4 as white crystals (0.0381 g, 40% yield from 1b, 0.1111
g, 30% yield from 1d). 1
H NMR (400 MHz; CD3OD) δ: 8.21 (1H, s, 8-H), 5.91 (1H, d,
J1′,2′ = 6.1 Hz, 1′-H), 4.67 (1H, apparent t, 2′-H), 4.32 (1H, m, 3′-H), 4.15 (1H, m, 4′-H),
3.89 (1H, dd, J5′a,4′ = 2.7 Hz, J5′a,b = 12.5 Hz, 5′a-H), 3.75 (1H, dd, J5′b,4′ = 3.0 Hz, J5′b,a
= 12.5 Hz, 5′b-H); 13
C NMR (101 MHz; CD3OD) δ: 157.17, 150.57, 141.59, 120.74,
120.45, 90.96, 87.84, 75.42, 72.40, 63.26.
Chapter 7
192
2-Iodo-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-adenine (4a)135
The title compound was prepared according to the general synthetic procedure C, from a
suspension of 4 (0.148 g, 1 equiv.) in acetone (10 mL), adding 70% HClO4 (82 µL) and
NH4OH (82 µL). The crude product was purified by silica gel chromatography with an
eluent 98:2 CH2Cl2/MeOH. 4a was obtained as a colourless oil (0.160 g, 98% yield). 1H
NMR (400 MHz; CD3OD) δ: 8.20 (1H, s, 8-H), 6.10 (1H, d, J1′,2′ = 3.2 Hz, 1′-H), 5.24
(1H, m, 2′-H), 5.01 (1H, m, 3′-H), 4.34 (1H, m, 4′-H), 3.79 (1H, dd, J5′a,4′ = 3.7 Hz, J5′a,b
= 12.0 Hz, 5′a-H), 3.72 (1H, dd, J5′b,4′ = 4.5 Hz, J5′b,a = 12.0 Hz, 5′b-H), 1.61 (3H, s,
isoprop.), 1.38 (3H, s, isoprop.); m/z (ESI) 432.0167 [M-H]-, C13H15IN5O4
- requires
432.0174.
2-Phenyl-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-adenine (4b)136
4a (0.083 g, 1 equiv.), phenyl boronic acid (1.7 equiv.), Pd(OAc)2 (0.12 equiv.), (2-
biphenyl)dicyclohexylphosphine (0.16 equiv.) and K3PO4 (2.2 equiv.) were added in a
sealed tube and purged with nitrogen for 10 min; after addition of anhydrous 1,4-
dioxane (3 mL), the tube was sealed and heated at 100 °C for 24 h. The reaction mixture
was evaporated to dryness in vacuo and the crude product was purified by silica gel
chromatography with a gradient 0-6% MeOH/CH2Cl2. 4b was obtained as white
Chapter 7
193
crystals (0.0318 g, 43% yield). 1H NMR (400 MHz; CDCl3) δ: 8.33 – 8.08 (2H, m, Ph),
7.83 (1H, s, 8-H), 7.46 (3H, m, Ph), 5.90 (1H, d, J1′,2′ = 4.6 Hz, 1′-H), 5.64 – 5.41 (1H,
m, 2′-H), 5.19 (1H, m, 3′-H), 4.48 (1H, m, 4′-H), 3.81 (2H, m, 5′-H2), 1.66 (3H, s,
isoprop.), 1.41 (3H, s, isoprop.); 13
C NMR (101 MHz; CDCl3) δ: 160.95, 155.71,
150.15, 140.86, 138.12, 130.27, 128.64, 128.47, 119.71, 114.30, 93.45, 85.92, 82.60,
81.71, 63.34, 27.76, 25.48; m/z (ESI) 384.1670 [M+H]+, C19H22N5O4
+ requires
384.1666.
2-(N-Me-Pyrrol-2-yl)-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-adenine (4c)
4a (0.083 g, 1 equiv.), tributyl-(N-Me-pyrrol-2-yl) stannane (5 equiv.), optional
anhydrous LiCl (1.2 equiv.), PdCl2(PPh3)2 (0.06 equiv.) were added in a sealed tube and
purged with nitrogen for 10 min; after addition of anhydrous 1,4-dioxane (2.5 mL), the
tube was sealed and heated at 90 °C for 48 h. The reaction mixture was evaporated to
dryness in vacuo and the crude product was purified by silica gel chromatography with
an eluent 98:2 CH2Cl2/MeOH. 4c was never isolated pure, but in a mixture with the
starting material 4a and stannane contamination; the 1H NMR showed 43-50%
conversion. 1H NMR (400 MHz; CD3OD) δ: 8.30 (1H, s, 8-H), 6.97 (1H, br s, 5-H
pyrrole), 6.86 (1H, br s, 3-H pyrrole), 6.20 (1H, d, J1′,2′ = 2.9 Hz, 1′-H), 6.16 – 6.11 (1H,
m, 4-H pyrrole), 5.39 (1H, m, 2′-H), 5.02 – 5.00 (1H, m, 3′-H), 4.09 (3H, s, N-CH3),
4.38 (1H, m, 4′-H), 3.87 – 3.67 (2H, m, 5′-H2), 1.7 – 1.3 (6H, 2s, 2 isoprop.).
Chapter 7
194
Phosphoric acid mono-5′-[2-iodo-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-
adenine] (4g)136
The title compound was prepared according to the general synthetic procedure D1, from
4a (0.0753 g, 1 equiv.) and POCl3 (0.1 mL, 6.2 equiv.) in anhydrous acetonitrile (2.5
mL). The crude product was purified by purification method 1 (0-60% MeOH against
0.05 M TEAB buffer over 350 mL, then isocratic 60% over 50 mL, flow: 3 mL/min,
fraction size: 5 mL), to give 4g as a colourless oil (0.0102 g, 1.45 equiv. TEA, 9%
yield). 1
H NMR (400 MHz; D2O) δ: 8.35 (1H, s, 8-H), 6.19 (1H, d, J1′,2′ = 3.2 Hz, 1′-H),
5.36 (1H, m, 2′-H), 5.17 (1H, m, 3′-H), 4.62 (1H, br s, 4′-H), 3.99 (3H, m, 5′-H2), 3.18
(8.7H, q, J = 7.3 Hz, CH2 TEA), 1.66 (3H, s, isoprop.), 1.43 (3H, s, isoprop.), 1.25
(13.2H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 1.54.
Phosphoric acid mono-5′-[2-phenyl-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-
adenine] (4h)136
The title compound was prepared according to the general synthetic procedure D1, from
4b (0.0155 g, 1 equiv.), proton sponge (6 equiv.) and POCl3 (25 μL, 6 equiv.) in
triethylphosphate (4 mL). The crude product was purified by purification method 2 (0-
100% 1 M TEAB buffer against H2O over 300 mL, flow: 2 mL/min, fraction size: 5
mL), to give 4h as a colourless oil (yield not calculated because the product was full of
TEA salts). 1
H NMR (400 MHz; D2O) δ: 8.33 (1H, s, 8-H), 8.06 (2H, m, Ph), 7.49 (3H,
Chapter 7
195
m, Ph), 6.32 (1H, br s, 1′-H), 5.44 (1H, s, 2′-H), 5.21 (1H, s, 3′-H), 4.61 (1H, s, 4′-H),
4.01 (2H, m, 5′-H2), 3.35 (85.7H, q, J = 7.3 Hz, CH2 TEA), 1.68 (3H, s, isoprop.), 1.45
(3H, s, isoprop.), 1.25 (88.7H, t, J = 7.2 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ:
0.23.
Phosphoric acid mono-5′-[2-iodo-9β-D-ribofuranosyl-adenine] (7a)
The title compound was prepared according to the general synthetic procedure E, from
4g (0.0102 g, 1 equiv.). The crude product was purified by purification method 1 (0-
50% MeOH against 0.05 M TEAB buffer over 300 mL, then isocratic 50% over 50 mL,
flow: 3 mL/min, fraction size: 5 mL), to obtain 7a as a colourless oil (0.0037 g, 1.86
equiv. TEA, 36% yield).
The title compound was prepared according to the general synthetic procedures D2 and
E. A mixture of 4a and 4c (0.0308 g, 1 equiv., 43% of 4c by 1H NMR estimation) was
dissolved in anhydrous CH2Cl2 (2 mL) and DIPEA (2.5 equiv.) and 2-cyanoethyl N,N-
diisopropyl chlorophosphoramidite (1 equiv.) were added via syringe. The reaction was
stirred under nitrogen at room temperature for 2 h. Then, 5.0 – 6.0 M tBuOOH in
nonane solution was added to the reaction and stirred for 1 h. After extractions, the
organic phase was evaporated to dryness in vacuo and the residue was suspended in
NH4OH under stirring at room temperature overnight. The solution was evaporated to
dryness in vacuo, suspended in water and stirred with Dowex 88 (H+) overnight. The
crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 350 mL, then isocratic 50% over 50 mL, flow: 3 mL/min, fraction
size: 5 mL), to obtain 7a as a colourless oil (0.005 g, 1.68 equiv. TEA, 19% yield).
Chapter 7
196
1H NMR (400 MHz; D2O) δ: 8.47 (1H, s, 8-H), 6.02 (1H, d, J1′,2′ = 5.7 Hz, 1′-H), 4.72
(1H, apparent t, 2′-H), 4.50 – 4.44 (1H, m, 3′-H), 4.34 (1H, m, 4′-H), 4.07 – 3.89 (2H,
m, 5′-H2), 3.17 (10.1H, q, J = 7.3 Hz, CH2 TEA), 1.25 (15.1H, t, J = 7.3 Hz, CH3 TEA);
13C NMR (101 MHz; D2O) δ: 156.10, 150.29, 145.45, 140.40, 120.35, 87.93, 85.06,
75.49, 71.26, 64.78, 47.49, 9.07; 31
P NMR (162 MHz; D2O) δ: 1.65; m/z (ESI)
473.9666 [M+H]+, C10H14IN5O7P
+ requires 473.9670.
Phosphoric acid mono-5′-[2-phenyl-9β-D-ribofuranosyl-adenine] (7b)
The title compound was prepared according to the general synthetic procedures D2 and
E. 4b (0.032 g, 1 equiv.) was dissolved in anhydrous CH2Cl2 (2.5 mL) and DIPEA (2.5
equiv.) and 2-cyanoethyl N,N-diisopropyl chlorophosphoramidite (2 equiv.) were added
via syringe. The reaction was stirred under nitrogen at room temperature for 2 h. The
intermediate 4e was isolated as a mixture of diastereomers by silica gel
chromatography, with an eluent 80:20 CH2Cl2/EtOAc with 3% TEA. 1H NMR (400
MHz; CDCl3) δ: 8.42 – 8.34 (2H, m, Ph), 7.98 (1H, s, 8-H), 7.49 – 7.41 (3H, m, Ph),
6.23 (0.5H, d, J1′,2′ = 2.1 Hz, 1′-H), 6.20 (0.5H, d, J1′,2′ = 2.1 Hz, 1′-H), 5.58 (0.5H, dd, J
= 6.2, 2.1 Hz, 2′-H), 5.55 (0.5H, dd, J = 6.2, 2.1 Hz, 2′-H), 5.21 – 5.15 (1H, m, 3′-H),
4.50 – 4.45 (1H, m, 4′-H), 3.96 – 3.66 (4H, m, 5′-H2, P-O-CH2), 3.51 (2H, m, N-CH),
2.66 (2H, m, CH2-CN), 1.65 (3H, s, isoprop.), 1.42 (3H, s, isoprop.), 1.14 – 1.05 (6H,
m, N-isoprop.); 31
P NMR (162 MHz; CDCl3) δ: 148.98. Then, 5.0 – 6.0 M tBuOOH in
nonane solution was added to the reaction and stirred for 1 h. After extractions, the
organic phase was evaporated to dryness in vacuo and the residue was suspended in
NH4OH under stirring at room temperature overnight. The solution was evaporated to
dryness in vacuo, suspended in water and stirred with Dowex 88 (H+) overnight. The
crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 350 mL, then isocratic 60% over 50 mL, flow: 3 mL/min, fraction
Chapter 7
197
size: 5 mL), to obtain 7b as a colourless oil (0.0075 g, 0.92 equiv. TEA, 17% yield). 1H
NMR (400 MHz; D2O) δ: 8.35 (1H, s, 8-H), 7.90 (2H, d, J = 7.0 Hz, Ph), 7.40 (3H, m,
Ph), 6.11 (1H, d, J1′,2′ = 5.2 Hz, 1′-H), 4.68 (1H, apparent t, 2′-H), 4.48 (1H, m, 3′-H),
4.32 (1H, br s, 4′-H), 4.17 – 3.98 (2H, m, 5′-H2), 3.13 (5.5H, q, J = 7.3 Hz, CH2 TEA),
1.22 (9.6H, t, J = 7.3 Hz, CH3 TEA); 13
C NMR (101 MHz; D2O) δ: 156.37, 152.78,
152.38, 145.28, 134.60, 132.37, 131.31, 130.25, 129.66, 99.29, 88.30, 84.94, 75.33,
71.21, 65.19, 59.67, 47.51, 9.07; 31
P NMR (162 MHz; D2O) δ: 0.94; m/z (ESI)
422.0870 [M-H]-, C16H17N5O7P
- requires 422.0871.
Phosphoric acid mono-5′-[2-(N-Me-pyrrol-2-yl)-9β-D-ribofuranosyl-adenine] (7c)
The title compound was prepared according to the general synthetic procedures D2 and
E. A mixture of 4a and 4c (0.0308 g, 1 equiv., 43% of 4c by 1H NMR estimation) was
dissolved in anhydrous CH2Cl2 (2 mL) and DIPEA (2.5 equiv.) and 2-cyanoethyl N,N-
diisopropyl chlorophosphoramidite (1 equiv.) were added via syringe. The reaction was
stirred under nitrogen at room temperature for 2 h. Then, 5.0 – 6.0 M tBuOOH in
nonane solution was added to the reaction and stirred for 1 h. After extractions, the
organic phase was evaporated to dryness in vacuo and the residue was suspended in
NH4OH under stirring at room temperature overnight. The solution was evaporated to
dryness in vacuo, suspended in water and stirred with Dowex 88 (H+) overnight. The
crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 350 mL, then isocratic 50% over 50 mL, flow: 3 mL/min, fraction
size: 5 mL), to obtain 7c as a yellow oil (0.0067 g, yield not calculated because of
stannane impurities still present in the product). 1H NMR (400 MHz; D2O) δ: 8.91 (1H,
s, 8-H), 8.21 (1H, d, J = 1.7 Hz, 5-H pyrrole), 7.62 (2H, d, 3-H pyrrole, 4-H pyrrole),
6.28 (1H, d, J1′,2′ = 5.4 Hz, 1′-H), 4.81 (1H, m, 2′-H), 4.55 – 4.48 (1H, m, 3′-H), 4.40
Chapter 7
198
(1H, br s, 4′-H), 4.14 (2H, m, 5′-H2), 3.98 – 3.86 (3H, m, N-CH3), 3.17 (5.2H, q, J = 7.3
Hz, CH2 TEA); 31
P NMR (162 MHz; D2O) δ: 0.54.
7.2.3 Preparation of 6-substituted AMP derivatives (8a-b)
6-Phenyl-9β-D-ribofuranosyl-purine (5a)147
The title compound was prepared according to the general synthetic procedure B1, from
5 (0.2 g, 1 equiv.), using Na2PdCl4, TPPTS, phenyl boronic acid and K2CO3 in water (6
mL) at 100 ºC. After 1 h, the reaction was cooled down and extracted with EtOAc; the
organic phase was dried over anhydrous MgSO4 and evaporated to dryness in vacuo to
obtain 5a as a white powder (0.1136 g, 50% yield). 1H NMR (400 MHz; DMSO-d6) δ:
9.02 (1H, s, 2-H), 8.93 (1H, s, 8-H), 8.84 (2H, d, J = 8.0 Hz, Ph), 7.55 – 7.65 (3H, m,
Ph), 6.11 (1H, d, J1′,2′ = 5.5 Hz, 1′-H), 5.56 (1H, d, J = 5.7 Hz, 2′-OH), 5.25 (1H, d, J =
5.0 Hz, 3′-OH), 5.13 (1H, t, J = 5.4 Hz, 5′-OH), 4.67 (1H, m, 2′-H), 4.23 (1H, m, 3′-H),
4.02 (1H, m, 4′-H), 3.76 – 3.59 (2H, m, 5′-H2).
Chapter 7
199
6-(Pyrrol-2-yl)-9β-D-ribofuranosyl-purine (5b)
The title compound was prepared according to the general synthetic procedure B1, from
5 (0.2 g, 1 equiv.), using Na2PdCl4, TPPTS, N-Boc-2-pyrrole boronic acid and K2CO3 in
water (6 mL) at 100 ºC. After 24 h, the reaction was cooled down and evaporated to
dryness in vacuo. The crude product was purified by purification method 1 (0-50%
MeOH against 0.05 M TEAB buffer over 350 mL, then isocratic 60% over 50 mL,
flow: 2 mL/min, fraction size: 5 mL) to obtain 5b as a yellow oil (0.0825 g, 37% yield).
1H NMR (400 MHz; CD3OD) δ: 8.69 (1H, s, 2-H), 8.59 (1H, s, 8-H), 7.51 (1H, dd, J5,4
= 3.8 Hz, J5,3 = 1.4 Hz, 5-H pyrrole), 7.16 (1H, dd, J3,4 = 2.5 Hz, J3,5 = 1.4 Hz, 3-H
pyrrole), 6.36 (1H, dd, J4,5 = 3.7 Hz, J4,3 = 2.6 Hz, 4-H pyrrole), 6.09 (1H, d, J1′,2′ = 6.0
Hz, 1′-H), 4.79 (1H, apparent t, 2′-H), 4.37 (1H, m, 3′-H), 4.19 (1H, m, 4′-H), 3.91 (1H,
dd, J5′a,b = 12.4 Hz, J5′a,4′ = 2.7 Hz, 5′a-H), 3.78 (1H, dd, J5′b,a = 12.4 Hz, J5′b,4′ = 2.9 Hz,
5′b-H), 2.80 (4H, q, J = 7.2 Hz, CH2 TEA), 1.14 (5.9H, t, J = 7.3 Hz, CH3 TEA); 13
C
NMR (101 MHz; CD3OD) δ: 153.04, 151.97, 149.51, 145.16, 129.65, 128.79, 125.20,
116.57, 111.86, 90.92, 87.88, 75.54, 72.46, 63.29, 47.30, 10.36; m/z (ESI) 316.1046
[M-H]-, C14H14N5O4
- requires 316.1051.
Chapter 7
200
6-Phenyl-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-purine (5c)
The title compound was prepared according to the general synthetic procedure C, from a
suspension of 5a (0.1136 g, 1 equiv.) in acetone (8 mL), adding 70% HClO4 (70 µL)
and NH4OH (70 µL). The mixture was extracted with EtOAc/brine (3x), and the organic
phase dried over anhydrous MgSO4; then, the solution was evaporated to dryness in
vacuo to give 5c as a colourless oil (0.109 g, 85% yield). 1H NMR (400 MHz; CD3OD)
δ: 8.95 (1H, s, 2-H), 8.73 (1H, s, 8-H), 8.70 – 8.61 (2H, m, Ph), 7.62 – 7.50 (3H, m,
Ph), 6.35 (1H, d, J1′,2′ = 3.0 Hz, 1′-H), 5.42 (1H, m, 2′-H), 5.09 (1H, m, 3′-H), 4.40 (1H,
m, 4′-H), 3.80 (1H, dd, J5′a,b = 12.0 Hz, J5′a,4′ = 3.8 Hz, 5′a-H), 3.73 (1H, dd, J5′b,a = 12.0
Hz, J5′b,4′ = 4.4 Hz, 5′b-H), 1.64 (3H, s, isoprop.), 1.40 (3H, s, isoprop.).
6-(Pyrrol-2-yl)-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-purine (5d)
The title compound was prepared according to the general synthetic procedure C, from a
suspension of 5b (0.0825 g, 1 equiv.) in acetone (8 mL), adding 70% HClO4 (56 µL)
and NH4OH (56 µL). The mixture was extracted with EtOAc/brine (3x), and the organic
phase dried over anhydrous MgSO4; then, the solution was evaporated to dryness in
Chapter 7
201
vacuo to give 5d as a yellow oil (0.0367 g, 39% yield). 1H NMR (400 MHz; CD3OD) δ:
8.71 (1H, s, 2-H), 8.59 (1H, s, 8-H), 7.51 (1H, dd, J5,4 = 3.8 Hz, J5,3 = 1.4 Hz, 5-H
pyrrole), 7.15 (1H, dd, J3,4 = 2.5 Hz, J3,5 = 1.4 Hz, 3-H pyrrole), 6.36 (1H, dd, J4,5 = 3.8
Hz, J4,3 = 2.6 Hz, 4-H pyrrole), 6.27 (1H, d, J1′,2′ = 3.2 Hz, 1′-H), 5.38 (1H, m, 2′-H),
5.08 (1H, m, 3′-H), 4.39 (1H, m, 4′-H), 3.80 (1H, dd, J5′a,b = 12.0 Hz, J5′a,4′ = 3.7 Hz,
5′a-H), 3.73 (1H, dd, J5′b,a = 12.0 Hz, J5′b,4′ = 4.3 Hz, 5′b-H), 1.63 (3H, s, isoprop.), 1.40
(3H, s, isoprop.).
Phosphoric acid mono-5′-[6-phenyl-9β-(2′,3′-O-isopropylidene-D-ribofuranosyl)-
purine] (5g)
The title compound was prepared according to the general synthetic procedure D1, from
5c (0.109 g, 1 equiv.) and POCl3 (0.1 mL, 3.6 equiv.) in anhydrous acetonitrile (3 mL).
The crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 300 mL, isocratic 50% over 50 mL, isocratic 60% over 50 mL, flow:
3 mL/min, fraction size: 5 mL), to give 5g as a colourless oil (0.024 g, 2.4 equiv. TEA,
12% yield). 5g was partially deprotected in the reaction conditions to give 8a as a
colourless oil (0.0448 g, 2.18 equiv. TEA, 24% yield). 1H NMR (400 MHz; D2O) δ:
8.80 (1H, s, 2-H), 8.75 (1H, s, 8-H), 8.07 (2H, m, Ph), 7.60 – 7.46 (3H, m, Ph), 6.27
(1H, d, J1′,2′ = 3.2 Hz, 1′-H), 5.36 (1H, m, 2′-H), 5.20 (1H, m, 3′-H), 4.64 (1H, br s, 4′-
H), 3.91 (2H, m, 5′-H2), 3.14 (14.4H, q, J = 7.3 Hz, CH2 TEA), 1.69 (3H, s, isoprop.),
1.45 (3H, s, isoprop.), 1.22 (21.8H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O)
δ: 3.73.
Chapter 7
202
Phosphoric acid mono-5′-[6-(pyrrol-2-yl)-9β-(2′,3′-O-isopropylidene-D-
ribofuranosyl)-purine] (5h)
The title compound was prepared according to the general synthetic procedure D1, from
5d (0.0179 g, 1 equiv.) and POCl3 (19 µL, 4 equiv.) in anhydrous acetonitrile (0.8 mL).
The reaction mixture was kept at 4 °C for 48 h. The crude product was purified by
purification method 1 (0-50% MeOH against 0.05 M TEAB buffer over 350 mL, then
isocratic 60% over 50 mL, flow: 3 mL/min, fraction size: 5 mL) to give 5h as a yellow
oil (0.0102 g, 1.8 equiv. TEA, 33% yield). 5h was partially deprotected in the reaction
conditions to give 8b as a yellow oil (0.0017 g, 1.8 equiv. TEA, 6% yield).1H NMR
(400 MHz; CD3OD) δ: 8.79 (1H, s, 2-H), 8.71 (1H, s, 8-H), 7.50 – 7.47 (1H, dd, J5,4 =
2.2 Hz, J5,3 = 1.3 Hz, 5-H pyrrole), 7.15 (1H, dd, J3,4 = 2.2 Hz, J3,5 = 1.3 Hz, 3-H
pyrrole), 6.35 (1H, m, 4-H pyrrole), 6.33 (1H, d, J1′,2′ = 3.2 Hz, 1′-H), 5.44 (1H, dd, J2′,3′
= 6.0 Hz, J1′,2′ = 3.2 Hz, 2′-H), 5.18 (1H, dd, J2′,3′ = 6.0 Hz, J3′,4′ = 1.8 Hz, 3′-H), 4.50
(1H, m, 4′-H), 4.07 – 3.96 (2H, m, 5′-H2), 3.03 (10.8H, q, J = 7.3 Hz, CH2 TEA), 1.63
(3H, s, isoprop.), 1.40 (3H, s, isoprop.), 1.22 (16.8H, t, J = 7.3 Hz, CH3 TEA); 31
P
NMR (162 MHz; CD3OD) δ: 1.65.
Chapter 7
203
Phosphoric acid mono-5′-[6-phenyl-9β-D-ribofuranosyl-purine] (8a)
The title compound was prepared according to the general synthetic procedure E, from
5g (0.024 g, 1 equiv.). The crude product was purified by purification method 1 (0-50%
MeOH against 0.05 M TEAB buffer over 300 mL, then isocratic 60% over 100 mL,
flow: 3 mL/min, fraction size: 5 mL), to obtain 8a as a colourless oil (0.0064 g, 2.1
equiv. TEA, 30% yield).
The title compound was prepared according to the general synthetic procedures D2 and
E. 5c (0.177 g, 1 equiv.) was dissolved in anhydrous CH2Cl2 (3 mL) and DIPEA (2.5
equiv.) and 2-cyanoethyl N,N-diisopropyl chlorophosphoramidite (1 equiv.) were added
via syringe. The reaction was stirred under nitrogen at room temperature for 1.5 h.
Then, 5.0 – 6.0 M tBuOOH in nonane solution was added to the reaction and stirred for
2 h. After extractions, the organic phase was evaporated to dryness in vacuo and the
phosphate intermediate isolated and purified as a mixture of diastereomers by silica gel
chromatography with an eluent 80:20 CH2Cl2/MeOH. 1H NMR (400 MHz; CDCl3) δ:
8.96 (1H, d, J = 1.1 Hz, 2-H), 8.73 – 8.65 (2H, m, Ph), 8.28 (0.5H, s, 8-H), 8.23 (0.5H,
s, 8-H), 7.54 – 7.43 (3H, m, Ph), 6.21 (0.5H, d, J1′,2′ = 2.3 Hz, 1′-H), 6.18 (0.5H, d, J1′,2′
= 2.3 Hz, 1′-H), 5.44 (0.5H, dd, J = 6.4, 2.3 Hz, 2′-H), 5.41 (0.5H, dd, J = 6.4, 2.0 Hz,
2′-H), 5.12 (1H, dt, J = 6.5, 3.4 Hz, 3′-H), 4.50 – 4.41 (1H, m, 4′-H), 4.20 – 3.93 (4H,
m, 5′-H2, P-O-CH2), 3.52 – 3.16 (4H, m, N-CH), 2.76 – 2.59 (2H, m, CH2-CN), 1.59
(3H, s, isoprop.), 1.34 (3H, s, isoprop.), 1.15 (12H, d, J = 13.6 Hz, N-isoprop.); 31
P
NMR (162 MHz; CDCl3) δ: 14.17. The residue was suspended in NH4OH under stirring
at room temperature overnight. The solution was evaporated to dryness in vacuo,
suspended in water and stirred with Dowex 88 (H+) overnight. After filtration, the crude
product was purified by purification method 1 (0-50% MeOH against 0.05 M TEAB
Chapter 7
204
buffer over 300 mL, then isocratic 60% over 100 mL, flow: 3 mL/min, fraction size: 5
mL), to give 8a as a colourless oil (0.0191 g, 1.37 equiv. TEA, 7% yield).
1H NMR (400 MHz; D2O) δ: 8.86 (1H, s, 2-H), 8.78 (1H, s, 8-H), 8.23 – 8.08 (2H, m,
Ph), 7.58 (3H, m, Ph), 6.25 (1H, d, J1′,2′ = 5.3 Hz, 1′-H), 4.68 (1H, m, 2′-H), 4.51 (1H,
m, 3′-H), 4.40 (1H, m, 4′-H), 4.14 (2H, m, 5′-H2), 3.17 (8.2H, q, J = 7.3 Hz, CH2 TEA),
1.25 (13.9H, t, J = 7.3 Hz, CH3 TEA); 13
C NMR (101 MHz; D2O) δ: 156.04, 152.66,
152.25, 145.22, 134.40, 132.36, 131.13, 130.14, 129.56, 88.31, 84.92, 75.37, 71.21,
59.70, 47.50, 23.08, 9.07, 8.23; 31
P NMR (162 MHz; D2O) δ: 2.98; m/z (ESI) 407.0754
[M-H]-, C16H16N4O7P
- requires 407.0762.
Phosphoric acid mono-5′-[6-(pyrrol-2-yl)-9β-D-ribofuranosyl-purine] (8b)
The title compound was prepared from 5h (0.0102 g, 1 equiv. TEA), dissolving it in an
acetic acid solution (AcOH/H2O, 1:9) at 90 °C for 3 h. After co-evaporation with EtOH
(3x) to remove AcOH traces, the crude product gave 8b as a colourless oil (0.0082 g,
complete conversion). 1H NMR (400 MHz; D2O) δ: 8.66 (1H, s, 2-H), 8.58 (1H, s, 8-H),
7.26 (1H, dd, J5,4 = 3.8 Hz, J5,3 = 1.4 Hz, 5-H pyrrole), 7.21 (1H, dd, J3,4 = 2.5 Hz, J5,3 =
1.4 Hz, 3-H pyrrole), 6.42 (1H, dd, J5,4 = 3.8 Hz, J3,4 = 2.6 Hz, 4-H pyrrole), 6.16 (1H,
d, J1′,2′ = 5.5 Hz, 1′-H), 4.75 (1H, m, 2′-H), 4.52 – 4.47 (1H, m, 3′-H), 4.40 – 4.35 (1H,
m, 4′-H), 4.18 – 4.06 (2H, m, 5′-H2); 13
C NMR (101 MHz; D2O) δ: 152.94, 151.51,
148.03, 144.01, 128.03, 127.34, 126.06, 115.89, 111.94, 88.04, 75.28, 71.24; 31
P NMR
(162 MHz; D2O) δ: 0.63; m/z (ESI) 396.0714 [M-H]-, C14H15N5O7P
- requires 396.0715.
Chapter 7
205
7.2.4 Preparation of 8-substituted AMP derivatives (9b-n)
Phosphoric acid mono-5′-[8-bromo-9β-D-ribofuranosyl-adenine] (9a)172
To a solution of adenosine monophosphate 3a (0.5 g, 1 equiv.) in aqueous KH2PO4 (50
mL, 0.25 M, pH 4), neat bromine (6.7 equiv.) was added drop-wise. The reaction was
stirred at room temperature in the dark overnight. The aqueous solution was washed
with CH2Cl2 until the colour remained pale yellow, and evaporated to dryness in vacuo.
The crude product was purified by purification method 1 (0-10% MeOH against 0.05 M
TEAB buffer over 400 mL, flow: 5 mL/min, fraction size: 5 mL) to give 9a as a
colourless oil (0.56 g, 2.2 equiv. TEA, 60% yield). 1H NMR (400 MHz; D2O) δ: 8.03
(1H, s, 2-H), 5.99 (1H, d, J1′,2′ = 6.0 Hz, 1′-H), 5.19 (1H, apparent t, 2′-H), 4.51 (1H, m,
3′-H), 4.25 – 3.93 (3H, m, 4′-H, 5′-H2), 3.10 (13.8H, q, J = 7.3 Hz, CH2 TEA), 1.19
(19.6H, t, J = 7.3 Hz, CH3 TEA); m/z (ESI) 423.9670 [M-H]-, C10H12BrN5O7P
- requires
423.9663; TLC: Rf 0.27 (tBuOH/1M TEAB buffer 6:4).
Phosphoric acid mono-5′-[8-phenyl-9β-D-ribofuranosyl-adenine] (9b)172
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0833 g, 1 equiv.) using Na2PdCl4 (0.025 equiv.), TPPTS (2.5 equiv. to Pd), phenyl
boronic acid (1.7 equiv.) and K2CO3 (1.5 equiv.) in water (2 mL) at 80 ºC. After 1 h, the
reaction was complete. The crude product was purified by purification method 1 (0-60%
Chapter 7
206
MeOH against 0.05 M TEAB buffer over 400 mL, flow: 3 mL/min, fraction size: 5 mL)
and 2 (0-60% 1 M TEAB buffer against H2O over 800 mL, flow: 4 mL/min, fraction
size: 5 mL) to give 9b as a colourless oil (0.0751 g, 1.6 equiv. TEA, 82% yield). 1H
NMR (400 MHz; D2O) δ: 8.18 (1H, s, 2-H), 7.68 – 7.49 (5H, m, Ph), 5.81 (1H, d, J1′,2′ =
6.2 Hz, 1′-H), 5.21 (1H, apparent t, 2′-H), 4.38 (1H, m, 3′-H), 4.01 (3H, m, 4′-H, 5′-H2),
2.97 (7.8H, q, J = 7.3 Hz, CH2 TEA), 1.20 (16.4H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR
(162 MHz; D2O) δ: 7.007; m/z (ESI) 422.0869 [M-H]-, C10H12BrN5O7P
- requires
422.0871; TLC: Rf 0.6 (iPA/H2O/NH4OH 6:3:1).
Phosphoric acid mono-5′-[8-(3-(N-Boc-aminomethyl)phenyl)-9β-D-ribofuranosyl-
adenine] (9c)172
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.074 g, 1 equiv.) using Na2PdCl4 (0.025 equiv.), TPPTS (2.5 equiv. to Pd), 3-(N-
Boc-aminomethyl)phenyl boronic acid (1.2 equiv.) and K2CO3 (3 equiv.) in water (2
mL) at 80 ºC. After 1 h, the reaction was complete. The crude product was purified by
purification method 1 (0-50% MeOH against 0.05 M TEAB buffer over 400 mL, flow:
3 mL/min, fraction size: 5 mL) to give 9c as a colourless oil (0.0762 g, 1.3 equiv. TEA,
78% yield). 1H NMR (400 MHz; CD3OD) δ: 8.18 (1H, s, 2-H), 7.68 – 7.63 (2H, m, Ph),
7.55 – 7.46 (2H, m, Ph), 5.83 (1H, d, J1′,2′ = 5.8 Hz, 1′-H), 5.51 (1H, apparent t, 2′-H),
4.49 (1H, m, 3′-H), 4.31 (2H, s, CH2NH2), 4.27 – 4.03 (3H, m, 4′-H, 5′-H2), 3.01 (8H, q,
J = 7.3 Hz, CH2 TEA), 1.42 (9H, s, CH3 Boc), 1.19 (12.1H, t, J = 7.3 Hz, CH3 TEA);
31P NMR (162 MHz; D2O) δ: 7.013; TLC: Rf 0.64 (
tBuOH/1M TEAB buffer 6:4).
Chapter 7
207
Phosphoric acid mono-5′-[8-(3-(aminomethyl)phenyl)-9β-D-ribofuranosyl-adenine]
(9d)
The conformer 9d(a) was prepared according to the general synthetic procedure B2,
from 9a (0.0712 g, 1 equiv.) using Na2PdCl4 (0.025 equiv.), TPPTS (2.5 equiv. to Pd),
3-(N-Boc-aminomethyl)phenyl boronic acid (1.2 equiv.) and K2CO3 (3 equiv.) in water
(6 mL) at 40 °C. After 2 days, the reaction seemed complete and the crude product was
purified by purification method 1 (0-8% MeOH against 0.05 M TEAB buffer over 350
mL, flow: 3 mL/min, fraction size: 5 mL) to give 9d(a) as a colourless oil (0.047 g, 0.5
equiv. TEA, 85% yield).
The two conformers, 9d(a) and 9d(b), of the title compound were prepared according to
the general synthetic procedure B2, from 9a (0.042 g, 1 equiv.) using Na2PdCl4 (0.025
equiv.), TPPTS (2.5 equiv. to Pd), 3-(aminomethyl)phenyl boronic acid (1.2 equiv.) and
K2CO3 (3 equiv.) in water (3 mL) at 80 °C. After 2 h, the reaction seemed complete and
the crude product was purified by purification method 1 (0-8% MeOH against 0.05 M
TEAB buffer over 350 mL, flow: 3 mL/min, fraction size: 5 mL) and 2 (0-60% 1 M
TEAB buffer against H2O over 700 mL, flow: 3 mL/min, fraction size: 5 mL).
The conformer 9d(a) was prepared treating 9c with TFA in MeOH under reflux (pH 1)
until to almost complete disappearance of the starting material. The mixture was
evaporated to dryness and the crude product purified by purification method 1 (0-8%
MeOH against 0.05 M TEAB buffer over 350 mL, flow: 3 mL/min, fraction size: 5
mL).
Conformer 9d(a): 1H NMR (400 MHz; D2O) δ: 8.24 (1H, s, 2-H), 7.75 – 7.58 (4H, m,
Ph), 5.82 (1H, d, J1′,2′ = 6.2 Hz, 1′-H), 5.17 (1H, apparent t, 2′-H), 4.36 (1H, m, 3′-H),
4.23 (2H, s, CH2NH2), 4.17 – 3.95 (3H, m, 4′-H, 5′-H2), 3.35 (8.1H, q, J = 7.3 Hz, CH2
Chapter 7
208
TEA), 1.23 (12.2H, t, J = 7.3 Hz, CH3 TEA); m/z (ESI) 451.1150 [M-H]-, C17H20N6O7P
-
requires 451.1137; TLC: Rf 0.52 (iPA/H2O/NH4OH 6:3:1).
Conformer 9d(b): 1H NMR (400 MHz; D2O) δ: 8.21 (1H, s, 2-H), 7.72 – 7.55 (4H, m,
Ph), 5.80 (1H, d, J1′,2′ = 6.4 Hz, 1′-H), 5.17 (1H, apparent t, 2′-H), 4.37 (1H, m, 3′-H),
4.21 (2H, d, CH2NH2), 4.19 – 3.98 (3H, m, 4′-H, 5′-H2), 1.22 (4.9H, t, J = 7.3 Hz CH3
TEA); m/z (ESI) 451.1130 [M-H]-, C17H20N6O7P
- requires 451.1137; TLC: Rf 0.81
(iPA/H2O/NH4OH 6:3:1).
Phosphoric acid mono-5′-[8-(pyridin-3-yl)-9β-D-ribofuranosyl-adenine] (9e)172
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0175 g, 1 equiv.) using Na2PdCl4 (0.02 equiv.), TPPTS (3 equiv. to Pd), 3-
pyridinyl boronic acid (1.2 equiv.) and K2CO3 (3 equiv.) in water (1 mL) at 100 ºC.
After 24 h, the reaction was complete. The crude product was purified by purification
method 1 (0-10% MeOH against 0.05 M TEAB buffer over 400 mL, flow: 3 mL/min,
fraction size: 5 mL) and 2 (0-25% 1 M TEAB buffer against H2O over 350 mL, flow: 2
mL/min, fraction size: 3 mL) to give 9e as a colourless oil (0.0123 g, 2 equiv. TEA,
70% yield). 1H NMR (400 MHz; D2O) δ: 8.82 (1H, br s, 2-H pyridine), 8.70 (1H, br s,
6-H pyridine), 8.22 (1H, s, 2-H), 8.18 (1H, m, 4-H pyridine), 7.64 (1H, br s, 5-H
pyridine), 5.77 (1H, d, 1′-H), 5.22 (1H, apparent t, 2′-H), 4.38 (1H, m, 3′-H), 4.27 – 3.88
(3H, m, 4′-H, 5′-H2), 3.14 (12.5H, q, J = 7.3 Hz, CH2 TEA), 1.21 (18.6H, t, J = 7.3 Hz,
CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 7.013; TLC: Rf 0.34 (tBuOH/1M TEAB buffer
6:4).
Chapter 7
209
Phosphoric acid mono-5′-[8-(pyridin-4-yl)-9β-D-ribofuranosyl-adenine] (9f)
The title compound was prepared according to general synthetic procedure B2, from 9a
(0.045 g, 1 equiv.) using Na2PdCl4 (0.09 equiv.), TPPTS (3 equiv. to Pd), 4-pyridinyl
boronic acid (1.5 equiv.) and K2CO3 (3.5 equiv.) in water (2 mL) at 100 °C. After 12 h,
the reaction seemed complete. The crude product was purified by purification method 1
(0-18% MeOH against 0.05 M TEAB buffer over 400 mL, flow: 3 mL/min, fraction
size: 5 mL) and 2 (0-60% 1 M TEAB buffer against H2O over 400 mL, flow: 2 mL/min,
fraction size: 5 mL) to give 9f as a colourless oil (0.0324 g, 1.58 equiv. TEA, 80%
yield). 1H NMR (400 MHz; D2O) δ: 8.72 (2H, br s, 2-H pyridine, 6-H pyridine), 8.27
(1H, s, 2-H), 7.78 (2H, br s, 3-H pyridine, 5-H pyridine), 5.86 (1H, d, J1′,2′ = 5.1 Hz, 1′-
H), 5.27 (1H, apparent t, 2′-H), 4.46 (1H, m, 3′-H), 4.14 (3H, m, 4′-H, 5′-H2), 3.17
(8.7H, q, J = 7.3 Hz, CH2 TEA), 1.27 (13.1H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162
MHz; D2O) δ: 7.019; m/z (ESI) 423.0821 [M-H]-, C15H16N6O7P
- requires 423.0824.
Phosphoric acid mono-5′-[8-(2,4-DMT-pyrimidin-5-yl)-9β-D-ribofuranosyl-
adenine] (9g)172
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0675 g, 1 equiv.) using Na2PdCl4 (0.025 equiv.), TPPTS (2.5 equiv. to Pd), 2,4-
DMT-5-pyrimidinyl boronic acid (1.2 equiv.) and K2CO3 (3 equiv.) in water (3 mL) at
80 ºC. After 1 h, the reaction was complete. The crude product was purified by
Chapter 7
210
purification method 1 (0-40% MeOH against 0.05 M TEAB buffer over 400 mL, flow:
3 mL/min, fraction size: 5 mL) to give 9g as a colourless oil (0.074 g, 1.4 equiv. TEA,
75% yield). 1H NMR (400 MHz; D2O) δ: 8.43 (1H, s, 6-H pyrimidine), 8.23 (1H, s, 2-
H), 5.56 (1H, d, J1′,2′ = 6.2 Hz, 1′-H), 5.19 (1H, apparent t, 2′-H), 4.44 (1H, m, 3′-H),
4.14 – 3.82 (9H, m, 4′-H, 5′-H2, 2 OCH3), 2.83 (8.4H, q, J = 7.3 Hz, CH2 TEA), 1.09
(12.3H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 7.000; m/z (ESI)
484.0976 [M-H]-, C16H19N7O9P
- requires 484.0987; TLC: Rf 0.35 (
tBuOH/1M TEAB
buffer 6:4).
Phosphoric acid mono-5′-[8-(pyrrol-2-yl)-9β-D-ribofuranosyl-adenine] (9h)172
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0175 g, 1 equiv.) using Na2PdCl4 (0.02 equiv.), TPPTS (3 equiv. to Pd), N-Boc-2-
pyrrole boronic acid (1.5 equiv.) and K2CO3 (3 equiv.) in water (1 mL) at 100 ºC. After
24 h, the reaction was complete. The crude product was purified by purification method
1 (0-12% MeOH against 0.05 M TEAB buffer over 150 mL, 12-40% over 200 mL,
flow: 3 mL/min, fraction size: 5 mL) and 2 (0-5% 1 M TEAB buffer against H2O over
50 mL, 5-28% over 75 mL, 28-40% over 20 mL, isocratic 40% over 20 mL, flow: 2
mL/min, fraction size: 5 mL), to give 9h as a colourless oil (0.0098 g, 2.2 equiv. TEA,
57% yield). 1H NMR (400 MHz; D2O) δ: 8.07 (1H, s, 2-H), 7.14 (1H, dd, J5,4 = 2.5 Hz,
J5,3 = 1.4 Hz, 5-H pyrrole), 6.79 (1H, dd, J3,4 = 3.7 Hz, J3,5 = 1.3 Hz, 3-H pyrrole), 6.33
(1H, dd, J4,3 = 3.5 Hz, J4,5 = 2.8 Hz, 4-H pyrrole), 6.05 (1H, d, J1′,2′ = 6.2 Hz, 1′-H), 5.19
(1H, apparent t, 2′-H), 4.48 (1H, m, 3′-H), 4.22 (1H, m, 4′-H), 4.06 (2H, m, 5′-H2), 3.14
(13.8H, q, J = 7.3 Hz, CH2 TEA), 1.20 (19.6H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162
MHz; D2O) δ: 7.032; m/z (ESI) 411.0812 [M-H]-, C14H16N6O7P
- requires 411.2871;
TLC: Rf 0.65 (tBuOH/1M TEAB buffer 6:4).
Chapter 7
211
Phosphoric acid mono-5′-[8-(furan-2-yl)-9β-D-ribofuranosyl-adenine] (9i)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0554 g, 1 equiv.) using Na2PdCl4 (0.09 equiv.), TPPTS (2.3 equiv. to Pd), 2-
furanyl boronic acid (1.5 equiv.) and K2CO3 (3 equiv.) in water (6 mL) at 100 °C. After
12 h, the reaction seemed complete. The crude product was purified by purification
method 1 (0-30% MeOH against 0.05 M TEAB buffer over 400 mL, flow: 3 mL/min,
fraction size: 5 mL) to give 9i as a colourless oil (0.031 g, 1.7 equiv. TEA, 67% yield).
1H NMR (400 MHz; D2O) δ: 8.10 (1H, s, 2-H), 7.69 (1H, br s, 5-H furan), 7.09 (1H, br
s, 3-H furan), 6.60 (1H, br s, 4-H furan), 6.20 (1H, d, J1′,2′ = 5.7 Hz, 1′-H), 5.31 (1H,
apparent t, 2′-H), 4.56 (1H, m, 3′-H), 4.13 (3H, m, 4′-H, 5′-H2), 3.12 (10.2H, q, J = 7.3
Hz, CH2 TEA), 1.21 (15.2H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ:
6.975; m/z (ESI) 412.0652 [M-H]-, C14H15N5O8P
- requires 412.0664.
Phosphoric acid mono-5′-[8-(1H-pyrazol-4-yl)-9β-D-ribofuranosyl-adenine] (9j)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.1068 g, 1 equiv.) using Na2PdCl4 (0.1 equiv.), TPPTS (2.5 equiv. to Pd), 1H-
Pyrazole-4-boronic acid (2 equiv.) and K3PO4 (1.5 equiv.) in water (10 mL) at 100 °C.
After 12 h, the reaction seemed complete. The crude product was purified by
purification method 1 (0-40% MeOH against 0.05 M TEAB buffer over 300 mL, then
Chapter 7
212
isocratic 45% over 100 mL, flow: 3 mL/min, fraction size: 5 mL) and 2 (0-100% 1 M
TEAB buffer against H2O over 300 mL, flow: 2 mL/min, fraction size: 5 mL) to give 9j
as a colourless oil (0.0569 g, 1.7 equiv. TEA, 57% yield). 1
H NMR (400 MHz; D2O) δ:
8.12 (1H, s, 2-H), 8.07 (2H, s, 3-H pyrazole, 5-H pyrazole), 5.90 (1H, d, J1′,2′ = 6.2 Hz,
1′-H), 5.25 (1H, apparent t, 2′-H), 4.48 – 4.43 (1H, m, 3′-H), 4.23 (1H, m, 4′-H), 4.11
(2H, m, 5′-H2), 3.15 (6.1H, q, J = 7.3 Hz, CH2 TEA), 1.23 (9.9H, t, CH3 TEA); 31
P
NMR (162 MHz; D2O) δ: 0.25; m/z (ESI) 412.0745 [M-H]-, C13H15N7O7P
- requires
412.0776.
Phosphoric acid mono-5′-[8-(5-formyl-thien-2-yl)-9β-D-ribofuranosyl-adenine]
(9k)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0554 g, 1 equiv.) using Na2PdCl4 (0.09 equiv.), TPPTS (2.3 equiv. to Pd), 5-
formyl-2-thienyl boronic acid (1.5 equiv.) and K2CO3 (3 equiv.) in water (6 mL) at 100
°C. After 12 h, the reaction was not finished, however the crude product was evaporated
and purified by purification method 1 (0-30% MeOH against 0.05 M TEAB buffer over
400 mL, flow: 3 mL/min, fraction size: 5 mL) and 2 (0-60% 1 M TEAB buffer against
H2O over 400 mL, flow: 2 mL/min, fraction size: 5 mL) to give 9k as a colourless oil
(0.0275 g, 1.6 equiv. TEA, 52% yield). 1H NMR (400 MHz; D2O) δ: 9.58 (1H, s, CHO),
8.05 (1H, s, 2-H), 7.69 (1H, s, 4-H thienyl), 7.53 (1H, 1s, 3-H thienyl), 5.89 (1H, d, 1′-
H), 5.30 (1H, apparent t, 2′-H), 4.53 (1H, m, 3′-H), 4.16 (3H, m, 4′-H, 5′-H2), 3.10
(9.6H, q, J = 7.3 Hz, CH2 TEA), 1.18 (14.6H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162
MHz; D2O) δ: 7.019; m/z (ESI) 423.0376 [M-H]-, C15H15N5O8PS
- requires 456.0384.
Chapter 7
213
Phosphoric acid mono-5′-[8-(5-bromo-thien-2-yl)-9β-D-ribofuranosyl-adenine] (9l)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0792 g, 1 equiv.) using Na2PdCl4 (0.1 equiv.), TPPTS (2.5 equiv. to Pd), 5-bromo-
2-thienyl boronic acid (0.5 equiv.) and K3PO4 (1.5 equiv.) in water (15 mL) at 100 °C.
After 12 h, the reaction seemed complete. The crude product was purified by
purification method 1 (0-40% MeOH against 0.05 M TEAB buffer over 250 mL, 40-
50% over 150 mL, flow: 3 mL/min, fraction size: 5 mL) to give 9l as a colourless oil
(0.015 g, 1.9 equiv. TEA, 17% yield). 1
H NMR (400 MHz; D2O) δ: 8.23 (1H, s, 2-H),
7.43 (1H, d, J3,4 = 4.0 Hz, 4-H thienyl), 7.24 (1H, d, J3,4 = 4.0 Hz, 3-H thienyl), 6.02
(1H, d, J1′,2′ = 6.1 Hz, 1′-H), 5.42 (1H, apparent t, 2′-H), 4.56 – 4.50 (1H, m, 3′-H), 4.28
– 4.20 (1H, m, 4′-H), 4.07 (2H, m, 5′-H2), 3.17 (11.2H, q, CH2 TEA), 1.25 (17.9H, t,
CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 2.73; m/z (ESI) 505.9554 [M-H]-,
C14H14BrN5O7PS- requires 505.9540.
Phosphoric acid mono-5′-[8-(indol-2-yl)-9β-D-ribofuranosyl-adenine] (9m)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.1068 g, 1 equiv.) using Na2PdCl4 (0.09 equiv.), TPPTS (3.5 equiv. to Pd), N-Boc-
indole-2-boronic acid (1.6 equiv.) and K2CO3 (3 equiv.) in water (6 mL) at 100 °C.
Chapter 7
214
After 12 h, the reaction seemed complete. The crude product was purified by
purification method 1 (0-45% MeOH against 0.05 M TEAB buffer over 500 mL, flow:
3 mL/min, fraction size: 5 mL) to give 9m as a colourless oil (0.0966 g, 2.2 equiv. TEA,
83% yield). 1
H NMR (400 MHz; D2O) δ: 7.88 (1H, s, 2-H), 7.53 (1H, d, J = 8.0 Hz, 6-H
indole), 7.33 (1H, d, J = 8.3 Hz, 9-H indole), 7.13 (1H, t, J = 7.5 Hz, 3-H indole), 7.02 –
6.94 (2H, m, 7-H indole, 8-H indole), 6.17 (1H, d, J1′,2′ = 5.7 Hz, 1′-H), 5.42 (1H,
apparent t, 2′-H), 4.55 (1H, m, 3′-H), 4.36 – 4.28 (1H, m, 4′-H), 4.09 (2H, m, 5′-H2),
3.12 (13.2H, q, J = 7.3 Hz, CH2 TEA), 1.21 (20.4H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR
(162 MHz; D2O) δ: 3.37; m/z (ESI) 461.0955 [M-H]-, C18H18N6O7P
- requires 461.0980.
Phosphoric acid mono-5′-[8-(benzofuran-2-yl)-9β-D-ribofuranosyl-adenine] (9n)
The title compound was prepared according to the general synthetic procedure B2, from
9a (0.0396 g, 1 equiv.) using Na2PdCl4 (0.09 equiv.), TPPTS (3 equiv. to Pd), 2-
benzofuranyl boronic acid (2 equiv.) and K2CO3 (2 equiv.) in water (3 mL) at 100 °C.
After 12 h, the conversion seemed at 50%. The crude product was purified by
purification method 1 (0-40% MeOH against 0.05 M TEAB buffer over 400 mL,
isocratic 50% over 50 mL, flow: 3 mL/min, fraction size: 5 mL) to give 9n as a
colourless oil (0.0038 g, 1.4 equiv. TEA, 10% yield). 1H NMR (400 MHz; D2O) δ: 7.89
(1H, s, 2-H), 7.40 (2H, m, 6-H benzofuran, 9-H benzofuran), 7.29 (1H, s, 3-H
benzofuran), 7.20 (1H, m, 7-H benzofuran), 7.04 (1H, m, 8-H benzofuran), 6.28 (1H, d,
J1′,2′ = 5.3 Hz, 1′-H), 5.39 (1H, apparent t, 2′-H), 4.63 (1H, m, 3′-H), 4.32 (1H, d, 4′-H),
4.27 – 4.06 (2H, m, 5′-H2), 3.15 (8.5H, q, J = 7.3 Hz, CH2 TEA), 1.23 (16.6H, t, J = 7.3
Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 0.82; m/z (ESI) 462.0848 [M-H]-,
C18H17N5O8P- requires 462.0820.
Chapter 7
215
7.2.5 Preparation of aryl/heteroaryl adenine-modified NAD+ derivatives (13a-b, 14a-b, 15b,c,e,g,h)
Morpholin-4-yl-phosphonic acid mono-5′-[2-iodo-9β-D-ribofuranosyl-adenine]
(10a)
The title compound was prepared according to the general synthetic procedure F, from
7a (0.009 g, 1 equiv.). The crude product was purified by purification method 1 (100%
0.05 M TEAB over 60 mL, 0-60% MeOH against 0.05 M TEAB buffer over 300 mL,
isocratic 60% over 100 mL, flow: 3 mL/min, fraction size: 5 mL) to give 10a as a
colourless oil (0.0089 g, 1.1 equiv. TEA, 97% yield). 1H NMR (400 MHz; D2O) δ: 8.30
(1H, s, 8-H), 6.01 (1H, d, J1′,2′ = 4.7 Hz, 1′-H), 4.76 (1H, m, 2′-H), 4.51 (1H, t, J = 4.9
Hz, 3′-H), 4.31 (1H, m, 4′-H), 4.08 – 3.93 (2H, m, 5′-H2), 3.53 (4H, m, CH2
morpholine), 3.17 (6.8H, q, J = 7.3 Hz, CH2 TEA), 2.88 (4H, m, CH2 morpholine), 1.24
(9.8H, t, CH3 TEA); 31
P NMR (162 MHz; D2O) δ: 7.37; m/z (ESI) 541.0105 [M-H]-,
C14H19IN6O7P- requires 541.0103.
Morpholin-4-yl-phosphonic acid mono-5′-[2-phenyl-9β-D-ribofuranosyl-adenine]
(10b)
The title compound was prepared according to the general synthetic procedure F, from
7b (0.0075 g, 1 equiv.). The crude product was purified by purification method 1 (100%
0.05 M TEAB over 100 mL, 0-50% MeOH against 0.05 M TEAB buffer over 250 mL,
Chapter 7
216
isocratic 60% over 50 mL, flow: 2.5 mL/min, fraction size: 5 mL) to give 10b as a
colourless oil (0.0086 g, 1.1 equiv. TEA, 98% yield). 1H NMR (400 MHz; D2O) δ: 8.28
(1H, s, 8-H), 8.01 (2H, m, Ph), 7.45 (3H, m, Ph), 6.10 (1H, d, J1′,2′ = 4.2 Hz, 1′-H), 4.87
(1H, apparent t, 2′-H), 4.63 (1H, apparent t, 3′-H), 4.30 (1H, m, 4′-H), 4.00 (2H, m, 5′-
H2), 3.37 (4H, m, CH2 morpholine), 2.98 (6.6H, q, J = 7.3 Hz, CH2 TEA), 2.76 (4H, m,
CH2 morpholine), 1.16 (10.1H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR (162 MHz; D2O) δ:
7.42; m/z (ESI) 491.1452 [M-H]-, C20H24N6O7P
- requires 491.1450.
Morpholin-4-yl-phosphonic acid mono-5′-[6-phenyl-9β-D-ribofuranosyl-purine]
(11a)
The title compound was prepared according to the general synthetic procedure F, from
8a (0.0191 g, 1 equiv.). The crude product was purified by purification method 1 (100%
0.05 M TEAB over 100 mL, 0-50% MeOH against 0.05 M TEAB buffer over 250 mL,
isocratic 60% over 50 mL, flow: 3 mL/min, fraction size: 5 mL) to give 11a as a
colourless oil (0.0172 g, 1.2 equiv. TEA, 82% yield). 1H NMR (400 MHz; D2O) δ: 8.78
(1H, s, 2-H), 8.68 (1H, s, 8-H), 8.05 (2H, m, Ph), 7.58 – 7.45 (3H, m, Ph), 6.16 (1H, d,
J1′,2′ = 4.6 Hz, 1′-H), 4.53 (1H, t, J = 4.9 Hz, 3′-H), 4.35 (1H, m, 4′-H), 4.12 – 3.96 (2H,
m, 5′-H2), 3.48 (4H, m, CH2 morpholine), 3.16 (7.4H, q, J = 7.3 Hz, CH2 TEA), 2.87
(4H, m, CH2 morpholine), 1.24 (11.6H, t, J = 7.2 Hz, CH3 TEA); 31
P NMR (162 MHz;
D2O) δ: 7.40; m/z (ESI) 476.1339 [M-H]-, C20H23N5O7P
- requires 476.1341.
Chapter 7
217
Morpholin-4-yl-phosphonic acid mono-5′-[6-(pyrrol-2-yl)-9β-D-ribofuranosyl-
purine] (11b)
The title compound was prepared according to the general synthetic procedure F, from
8b (0.015 g, 1 equiv.). The crude product was purified by purification method 1 (100%
0.05 M TEAB over 100 mL, 0-50% MeOH against 0.05 M TEAB buffer over 250 mL,
isocratic 60% over 50 mL, flow: 3 mL/min, fraction size: 5 mL) to give 11b as a
colourless oil (0.0132 g, 1.1 equiv. TEA, 88% yield). 1H NMR (400 MHz; D2O) δ: 8.54
(1H, s, 2-H), 8.52 (1H, s, 8-H), 7.19 (2H, m, 5-H pyrrole, 3-H pyrrole), 6.39 (1H, s, 4-H
pyrrole), 6.09 (1H, d, J1′,2′ = 4.6 Hz, 1′-H), 4.52 (1H, t, J = 4.9 Hz, 3′-H), 4.34 (1H, m,
4′-H), 4.11 – 3.92 (2H, m, 5′-H2), 3.49 (4H, m, CH2 morpholine), 3.15 (6.3H, q, J = 7.3
Hz, CH2 TEA), 2.85 (4H, m, CH2 morpholine), 1.23 (10.1H, t, J = 7.3 Hz, CH3 TEA);
31P NMR (162 MHz; D2O) δ: 7.41.
Morpholin-4-yl-phosphonic acid mono-5′-[8-phenyl-9β-D-ribofuranosyl-adenine]
(12b)112
The title compound was prepared according to the general synthetic procedure F, from
9b (0.0092 g, 1 equiv.) by NaI precipitation. 12b was obtained as sodium salt in
quantitative yield. 1H NMR (400 MHz; D2O) δ: 8.23 (1H, s, 2-H), 7.74 – 7.46 (5H, m,
Ph), 5.86 (1H, d, J1′,2′ = 5.3 Hz, 1′-H), 5.30 (1H, apparent t, 2′-H), 4.58 (1H, m, 3′-H),
Chapter 7
218
4.17 – 3.95 (3H, m, 4′-H, 5′-H2), 3.46 (4H, m, CH2 morpholine), 2.82 (4H, m, CH2
morpholine); m/z (ESI) 491.1452 [M-Na]-, C20H24N6O7P
- requires 491.1450.
Morpholin-4-yl-phosphonic acid mono-5′-[8-(3-(N-Boc-aminomethyl)phenyl)-9β-
D-ribofuranosyl-adenine] (12c)
The title compound was prepared according to the general synthetic procedure F, from
9c (0.0425 g, 1 equiv.). The crude product was purified by purification method 1 (0-
50% MeOH against 0.05 M TEAB buffer over 300 mL, then 50-60% over 100 mL,
flow: 3 mL/min, fraction size: 5 mL) to give 12c as a colourless oil (1.3 equiv. TEA,
quantitative yield). 1
H NMR (400 MHz; CD3OD) δ: 8.26 (1H, s, 2-H), 7.89 – 7.48 (4H,
m, Ph), 5.86 (1H, d, 1′-H), 5.58 (1H, apparent t, 2′-H), 4.60 (1H, m, 3′-H), 4.51 – 3.85
(5H, m, 4′-H, 5′-H2, CH2NHBoc), 3.55 (4H, m, CH2 morpholine), 2.91 (4H, m, CH2
morpholine), 3.06 (8H, q, J = 7.3 Hz, CH2 TEA), 1.45 (9H, s, CH3 Boc), 1.24 (11.8H, t,
J = 7.3 Hz, CH3 TEA).
Morpholin-4-yl-phosphonic acid mono-5′-[8-(pyridin-3-yl)-9β-D-ribofuranosyl-
adenine] (12e)
The title compound was prepared according to the general synthetic procedure F, from
9e (0.0123 g, 1 equiv.) by NaI precipitation. 12e was obtained as sodium salt in
quantitative yield. 1H NMR (400 MHz; D2O) δ: 8.84 (1H, m, 2-H pyridine), 8.71 (1H,
Chapter 7
219
m, 6-H pyridine), 8.20 (1H, s, 2-H), 8.19 – 8.04 (1H, m, 5-H pyridine), 5.81 (1H, d, 1′-
H), 5.38 (1H, m, 2′-H), 4.62 (1H, m, 3′-H), 4.26 – 3.90 (3H, m, 4′-H, 5′-H2), 3.48 (4H,
m, CH2 morpholine), 2.84 (4H, m, CH2 morpholine).
Morpholin-4-yl-phosphonic acid mono-5′-[8-(pyrrol-2-yl)-9β-D-ribofuranosyl-
adenine] (12h)
The title compound was prepared according to the general synthetic procedure F, from
9h (0.0417 g, 1 equiv.) by purification method 1 (0-50% MeOH against 0.05 M TEAB
buffer over 300 mL, then 50-60% over 100 mL, flow: 3 mL/min, fraction size: 5 mL) to
give 12h as a colourless oil (1.3 equiv. TEA, quantitative yield). 1H NMR (400 MHz;
D2O) δ: 8.16 (1H, s, 2-H), 7.13 (1H, s, 5-H pyrrole), 6.82 (1H, s, 3-H pyrrole), 6.35
(1H, s, 4-H pyrrole), 6.08 (1H, br s, 1′-H), 5.44 (1H, br s, 2′-H), 4.42 – 3.89 (4H, m, 3′-
H, 4′-H, 5′-H2), 3.36 (4H, m, CH2 morpholine), 2.72 (4H, m, CH2 morpholine), 3.15
(7.6H, q, J = 7.3 Hz, CH2 TEA), 1.21 (11.4H, t, J = 7.3 Hz, CH3 TEA).
Chapter 7
220
P1-(2-Iodo-adenine-9β-D-ribofuranos-5′-yl)-P
2-(nicotinamide-1β-D-ribofuranos-5′-
yl) pyrophosphate (13a)
The title compound was prepared according to the general synthetic procedure G, from
10a (0.0134 g, 1 equiv.) and β-NMN (2.4 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 360 mL, isocratic 50% over 40 mL, flow: 3 mL/min, fraction size: 5
mL) and treated with Chelex to give 13a as a colourless oil (0.0093 g, 0.2 equiv. TEA,
53% yield). 1H NMR (400 MHz; D2O) δ: 9.31 (1H, s, 2N-H), 9.06 (1H, s, 6N-H), 8.87
(1H, m, 4N-H), 8.31 (1H, s, 8-H), 8.21 (1H, m, 5N-H), 6.02 (1H, d, J1′′,2′′ = 5.0 Hz, 1′′-
H), 5.91 (1H, d, J1′,2′ = 5.3 Hz, 1′-H), 4.74 (1H, br s, 2′-H), 4.54 – 3.98 (9H, m, 2′′-H, 3′-
H, 3′′-H, 4′-H, 4′′-H, 5′-H2, 5′′-H2), 1.24 (1.8H, t, J = 7.3 Hz, CH3 TEA); 13
C NMR (126
MHz; D2O) δ: 156.87, 147.44, 143.83, 141.65, 138.06, 130.41, 121.21, 101.62, 98.90,
88.71, 88.27, 85.50, 79.19, 75.49, 72.29, 71.99, 67.03, 66.57, 60.58, 9.07; 31
P NMR
(162 MHz; D2O) δ: -11.4, -11.2; m/z (ESI) 787.9967 [M-2H]-, C21H25IN7O14P2
- requires
787.9985.
Chapter 7
221
P1-(2-Phenyl-adenine-9β-D-ribofuranos-5′-yl)-P
2-(nicotinamide-1β-D-ribofuranos-
5′-yl) pyrophosphate (13b)
The title compound was prepared according to the general synthetic procedure G, from
10b (1.7 equiv.) and β-NMN (0.0234 g, 1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 400 mL, flow: 3 mL/min, fraction size: 5 mL) to give 13b as a
colourless oil (0.017 g, 1.18 equiv. TEA, 28% yield). 1
H NMR (400 MHz; D2O) δ: 9.04
(1H, br s, 2N-H), 8.83 (1H, br s, 6N-H), 8.61 (1H, m, 4N-H), 8.09 – 7.86 (3H, m, 5N-H,
8-H, Ph), 7.45 (4H, m, Ph), 6.08 (1H, br s, 1′′-H), 5.76 (1H, br s, 1′-H), 4.84 (1H, br s,
2′-H), 4.56 (1H, br s, 2′′-H), 4.43 – 4.01 (6H, m, 3′-H, 3′′-H, 4′-H, 4′′-H, 5′-H2, 5′′-H2),
3.16 (7H, q, J = 7.2 Hz, CH2 TEA), 1.22 (10.6H, t, J = 7.2 Hz, CH3 TEA); 13
C NMR
(101 MHz; D2O) δ: 146.30, 142.68, 140.55, 138.29, 137.55, 137.23, 135.59, 131.55,
131.52, 129.56, 129.49, 128.80, 100.61, 93.30, 87.75, 84.39, 78.37, 74.32, 71.26, 65.04,
53.17, 52.17, 47.51, 9.07; 31
P NMR (162 MHz; D2O) δ: -11.5, -11.7; m/z (ESI)
738.1324 [M-H]-, C27H30N7O14P2
- requires 738.1331.
Chapter 7
222
P1-(6-Phenyl-purine-9β-D-ribofuranos-5′-yl)-P
2-(nicotinamide-1β-D-ribofuranos-
5′-yl) pyrophosphate (14a)
The title compound was prepared according to the general synthetic procedure G, from
11a (0.019 g, 1 equiv.) and β-NMN (1.8 equiv.). The reaction was stirred overnight. The
crude product was purified twice by purification method 1 (0-50% MeOH against 0.05
M TEAB buffer over 400 mL, flow: 3 mL/min, fraction size: 5 mL) and treated with
Chelex to give 14a as a colourless oil (0.0171 g, 0.4 equiv. TEA, 56% yield). 1H NMR
(400 MHz; D2O) δ: 9.18 (1H, s, 2N-H), 9.01 (1H, d, J6,5 = 5.8 Hz, 6N-H), 8.82 (1H, s, 2-
H), 8.78 (1H, s, 8-H), 8.66 (1H, d, J4,5 = 8.0 Hz, 4N-H), 8.11 (2H, d, J = 7.3 Hz, Ph),
8.06 – 8.01 (1H, m, 5N-H), 7.57 (3H, m, Ph), 6.21 (1H, d, J1′′,2′′ = 5.4 Hz, 1′′-H), 5.88
(1H, d, J = 4.2 Hz, 1′-H), 4.55 – 4.09 (10H, m, 2′-H, 2′′-H, 3′-H, 3′′-H, 4′-H, 4′′-H, 5′-
H2, 5′′-H2), 3.24 (2.4H, q, J = 7.3 Hz, CH2 TEA), 1.22 (4.1H, t, J = 7.2 Hz, CH3 TEA).
13C NMR (126 MHz; D2O) δ: 162.12, 156.90, 153.67, 153.43, 147.40, 146.20, 144.11,
141.19, 138.05, 135.24, 133.34, 132.02, 131.14, 131.00, 130.54, 130.10, 101.56, 98.89,
88.75, 85.62, 79.13, 75.74, 72.09, 67.00, 66.49, 60.58, 48.24, 9.07; 31
P NMR (162
MHz; D2O) δ: -11.3 (d, JP,P = 20.3 Hz), -11.7 (d, JP,P = 20.7 Hz); m/z (ESI) 723.1210
[M-2H]-, C27H29N6O14P2
- requires 723.1222.
Chapter 7
223
P1-[6-(Pyrrol-2-yl)-purine-9β-D-ribofuranos-5′-yl]-P
2-[nicotinamide-1β-D-
ribofuranos-5′-yl] pyrophosphate (14b)
The title compound was prepared according to the general synthetic procedure G, from
11b (0.0084 g, 1 equiv.) and β-NMN (2.1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-50% MeOH against 0.05 M
TEAB buffer over 400 mL, flow: 3 mL/min, fraction size: 5 mL) to give 14b as a
colourless oil (0.0132 g, 1.3 equiv. TEA, 87% yield). 1H NMR (400 MHz; D2O) δ: 9.11
(1H, s, 2N-H), 8.94 (1H, d, J6,5 = 5.1 Hz, 6N-H), 8.69 (1H, s, 2-H), 8.61 (1H, s, 8-H),
8.57 (1H, d, J = 7.7 Hz, 4N-H), 7.90 (1H, m, 5N-H), 7.23 (1H, s, 5-H pyrrole), 7.20 (1H,
d, J3,4 = 3.1 Hz, 3-H pyrrole), 6.43 (1H, m, 4-H pyrrole), 6.17 (1H, d, J1′′,2′′ = 5.8 Hz, 1′′-
H), 5.91 (1H, d, J1′,2′ = 4.4 Hz, 1′-H), 4.60 – 4.12 (10H, m, 2′-H, 2′′-H, 3′-H, 3′′-H, 4′-H,
4′′-H, 5′-H2, 5′′-H2), 3.18 (6.2H, q, J = 7.3 Hz, CH2 TEA), 1.25 (12H, t, J = 7.3 Hz, CH3
TEA); 13
C NMR (126 MHz; D2O) δ: 153.21, 151.85, 147.85, 146.39, 144.16, 143.33,
143.31, 140.35, 129.02, 127.81, 127.11, 126.18, 116.04, 112.16, 101.03, 98.14, 87.89,
87.59, 85.07, 78.47, 74.90, 71.57, 71.37, 59.83, 47.50, 9.07; 31
P NMR (162 MHz; D2O)
δ: -11.4, -11. 6; m/z (ESI) 714.1311 [M]+, C25H30N7O14P2
+ requires 714.1320.
Chapter 7
224
P1-(8-Phenyl-adenine-9β-D-ribofuranos-5′-yl)-P
2-(nicotinamide-1β-D-ribofuranos-
5′-yl) pyrophosphate (15b)112, 172
The title compound was prepared according to the general synthetic procedure G, from
12b (1.2 equiv.) and β-NMN (0.0038 g, 1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0% MeOH over 50 mL, 0-
40% MeOH against 0.05 M TEAB buffer over 350 mL, flow: 3 mL/min, fraction size: 5
mL) to give 15b as a colourless oil (0.004 g, 1.18 equiv. TEA, 41% yield). 1H NMR
(400 MHz; D2O) δ: 9.21 (1H, s, 2N-H), 9.06 (1H, d, J6,5 = 6.4 Hz, 6N-H), 8.68 (1H, m,
J4,5 = 8.1 Hz, J4.6 = 1.3 Hz, J4,2 = 1.5 Hz, 4N-H), 8.18 (1H, s, 2-H), 8.16 – 8.10 (1H, m,
5N-H), 7.65 – 7.50 (5H, m, Ph), 5.90 (1H, d, J1′′,2′′ = 5.1 Hz, 1′′-H), 5.79 (1H, d, J1′,2′ =
5.9 Hz, 1′-H), 5.14 (1H, apparent t, 2′-H), 4.43 – 4.06 (9H, m, 2′′-H, 3′-H, 3′′-H, 4′-H,
4′′-H, 5′-H2, 5′′-H2), 3.14 (6.8H, q, J = 7.3 Hz, CH2 TEA), 1.22 (10.6H, t, J = 7.3 Hz,
CH3 TEA); 13
C NMR (150.90 MHz; D2O) δ: 154.05, 153.70, 153.30, 150.91, 146.55,
143.18, 140.61, 132.04, 130.39, 130.25, 129.92, 129.63, 128.46, 119.14, 100.88, 89.82,
89.61, 87.85, 83.78, 78.41, 71.32, 71.23, 71.14, 70.38, 70.23, 66.61, 65.69, 59.73,
47.51, 9.07; 31
P NMR (162 MHz; D2O) δ: -11.7 (d, JP,P = 20.6 Hz); -11.3 (d, JP,P = 20.6
Hz); m/z (ESI) 738.1332 [M-2H]-, C27H30N7O14P2
- requires 738.1331; HPLC: retention
time 10.34 min (79.4%); TLC: Rf 0.43 (iPA/H2O/NH4OH 6:3:1).
Chapter 7
225
P1-{8-[3-(N-Boc-aminomethyl)phenyl]-adenine-9β-D-ribofuranos-5′-yl}-P
2-
{nicotinamide-1β-D-ribofuranos-5′-yl} pyrophosphate (15c)172
The title compound was prepared according to the general synthetic procedure G, from
12c (1.2 equiv.) and β-NMN (0.0176 g, 1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-10% MeOH against 0.05 M
TEAB buffer over 190 mL, 10-30% over 50 mL, isocratic 30% over 50 mL, flow: 3
mL/min, fraction size: 5 mL) to give 15c as a colourless oil (0.0331 g, 1.7 equiv. TEA,
61% yield) and treated with Chelex to obtain the sodium salt. 1H NMR (400 MHz; D2O)
δ: 9.16 (1H, s, 2N-H), 8.99 (1H, br s, 6N-H), 8.60 (1H, m, 4N-H), 8.07 (2H, m, 2-H, 5N-
H), 7.54 – 7.26 (4H, m, Ph), 5.85 (1H, br s, 1′′-H), 5.71 (1H, br s, 1′-H), 5.16 (1H, br s,
2′-H), 4.49 – 3.77 (11H, m, 2′′-H, 3′-H, 3′′-H, 4′-H, 4′′-H, 5′-H2, 5′′-H2, CH2NH2), 1.33
(9H, s, CH3 Boc); 13
C NMR (150.90 MHz; D2O) δ: 158.95, 155.55, 153.21, 152.45,
150.65, 148.28, 146.33, 142.95, 140.67, 140.41, 137.07, 134.10, 130.20 – 128.23,
124.83, 118.92, 101.81, 100.66, 96.99, 89.71, 87.67, 83.67, 81.92, 78.22, 75.81, 71.34 –
70.15, 66.34, 65.39, 44.10, 28.36; 31
P NMR (162 MHz; D2O) δ: -11.3 (d, JP,P = 20.6
Hz); -11.1 (d, JP,P = 20.6 Hz); m/z (ESI) 867.2120 [M-2H]-, C33H41N8O16P2
- requires
867.2121; HPLC: retention time 17.96 min (99.5%); TLC: Rf 0.75 (iPA/H2O/ NH4OH
6:3:1).
Chapter 7
226
P1-[8-(Pyridin-3-yl)-adenine-9β-D-ribofuranos-5′-yl]-P
2-[nicotinamide-1β-D-
ribofuranos-5′-yl] pyrophosphate (15e)172
The title compound was prepared according to the general synthetic procedure G, from
12e (0.0143 g, 1 equiv.) and β-NMN (1.1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-10% MeOH against 0.05 M
TEAB buffer over 300 mL, then 10-12% MeOH over 100 mL, flow: 3 mL/min, fraction
size: 5 mL) to give 15e as a colourless oil (0.0068 g, 1.18 equiv. TEA, 24% yield). 1H
NMR (400 MHz; D2O) δ: 9.29 (1H, s, 2N-H), 9.14 (1H, d, J6,5 = 6.0 Hz, 6N-H), 9.01 –
8.65 (3H, m, 4N-H, 2-H pyridine, 6-H pyridine), 8.34 – 8.13 (3H, m, 2-H, 5N-H, 4-H
pyridine), 7.70 (1H, br s, 5-H pyridine), 6.01 (1H, d, J1′′,2′′ = 5.1 Hz, 1′′-H), 5.80 (1H, d,
J1′,2′ = 5.7 Hz, 1′-H), 5.24 (1H, apparent t, 2′-H), 4.51 – 4.11 (9H, m, 2′′-H, 3′-H, 3′′-H,
4′-H, 4′′-H, 5′-H2, 5′′-H2), 3.19 (6.9H, q, J = 7.3 Hz, CH2 TEA), 1.27 (10.6H, t, J = 7.3
Hz, CH3 TEA); 13
C NMR (150.90 MHz; D2O) δ: 155.89, 153.53, 151.01, 150.53,
149.62, 146.60, 143.28, 140.56, 139.32, 134.53, 129.60, 119.50, 100.89, 89.71, 87.92,
84.08, 78.44, 71.38, 70.28, 66.41, 65.65, 47.52, 9.07; m/z (ESI) 739.1300 [M-2H]-,
C26H29N8O14P2- requires 739.1284; HPLC: retention time 7.07 min (95.8%); TLC: Rf
0.62 (iPA/H2O/NH4OH 6:3:1).
Chapter 7
227
P1-[8-(2,4-DMT-pyrimidin-5-yl)-adenine-9β-D-ribofuranos-5′-yl]-P
2-[nicotinamide-
1β-D-ribofuranos-5′-yl] pyrophosphate (15g)172
The title compound was prepared according to the general synthetic procedure G, from
12g (1.5 equiv.) and β-NMN (0.0112 g, 1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-22% MeOH against 0.05 M
TEAB buffer over 300 mL, 22-40% over 50 mL, isocratic 40% over 50 mL, flow: 3
mL/min, fraction size: 5 mL) to give 15g as a colourless oil (0.0139 g, 1.7eq TEA, 43%
yield) and treated with Chelex to obtain the sodium salt. 1H NMR (400 MHz; D2O) δ:
9.25 (1H, s, 2N-H), 9.08 (1H, m, 6N-H), 8.68 (1H, m, 4N-H), 8.33 (1H, m, 6-H
pyrimidine), 8.10 (2H, br s, 2-H, 5N-H), 6.01 (1H, d, J1′′,2′′ = 5.4 Hz, 1′′-H), 5.49 (1H, d,
J1′,2′ = 5.5 Hz, 1′-H), 5.08 (1H, apparent t, 2′-H), 4.53 – 3.85 (15H, m, 2′′-H, 3′-H, 3′′-H,
4′-H, 4′′-H, 5′-H2, 5′′-H2, 2 OCH3); 13
C NMR (150.90 MHz; D2O) δ: 169.85, 167.00,
161.02, 155.64, 153.60, 152.48, 150.47, 148.31, 146.40, 143.03, 137.11, 134.25,
129.37, 124.87, 119.49, 104.96, 101.81, 100.69, 97.02, 90.06, 87.71, 83.85, 81.78,
78.28, 75.82, 71.78 – 70.43, 66.90 – 65.50, 59.66, 56.29, 55.60; 31
P NMR (162 MHz;
D2O) δ: -11.3 (d, JP,P = 20.6 Hz); -11.1 (d, JP,P = 20.6 Hz); m/z (ESI) 800.1447 [M-2H]-
, C27H32N9O16P2- requires 800.1448; HPLC: retention time 11.41 min (96.5%); TLC: Rf
0.69 (iPA/H2O/ NH4OH 6:3:1).
Chapter 7
228
P1-[8-(Pyrrol-2-yl)-adenine-9β-D-ribofuranos-5′-yl]-P
2-[nicotinamide-1β-D-
ribofuranos-5′-yl] pyrophosphate (15h)172
The title compound was prepared according to the general synthetic procedure G, from
12h (1.5 equiv.) and β-NMN (0.0332 g, 1 equiv.). The reaction was stirred overnight.
The crude product was purified by purification method 1 (0-10% MeOH against 0.05 M
TEAB buffer over 400 mL, flow: 3 mL/min, fraction size: 5 mL) and 2 (0-80% 1 M
TEAB buffer against H2O over 400 mL, flow: 3 mL/min, fraction size: 5 mL), to give
15h as a colourless oil (0.0245 g, 1.01 equiv. TEA, 30% yield). 1H NMR (400 MHz;
D2O) δ: 9.09 (1H, s, 2N-H), 8.92 (1H, d, J5,6 = 6.2 Hz, 6N-H), 8.60 (1H, d, J4,5 = 8.1 Hz,
4N-H), 8.09 (1H, s, 2-H), 8.05 (1H, dd, J5,6 = 6.5 Hz, J5,4 = 7.8 Hz, 5N-H), 7.05 (1H, dd,
J5,4 = 2.4 Hz, J5,3 = 1.2 Hz, 5-H pyrrole), 6.57 (1H, dd, J3,4 = 3.6 Hz, J3,5 = 1.1 Hz, 3-H
pyrrole), 6.24 (1H, dd, J3,4 = 3.4 Hz, J5,4 = 2.9 Hz, 4-H pyrrole), 5.91 (1H, d, J1′′,2′′ = 6.1
Hz, 1′′-H), 5.73 (1H, d, J1′,2′ = 4.8 Hz, 1′-H), 5.33 (1H, apparent t, 2′-H), 4.54 (1H, dd, J
= 4.56 and 6.1 Hz, 3′-H), 4.41 – 3.98 (8H, m, 2′′-H, 3′′-H, 4′-H, 4′′-H, 5′-H2, 5′′-H2),
3.15 (6.08H, q, J = 7.3 Hz, CH2 TEA), 1.24 (9.14H, t, J = 7.3 Hz, CH3 TEA); 13
C NMR
(150.90 MHz; D2O) δ: 154.89, 151.98, 150.90, 147.10, 146.40, 142.91, 140.48, 134.14,
129.64, 124.27, 119.43, 113.82, 110.00, 100.81, 89.46, 87.79, 87.73, 83.81, 83.76,
78.41, 71.24, 70.89, 70.02, 66.47, 65.61, 59.60, 47.52, 9.07; 31
P NMR (162 MHz; D2O)
δ: -11.9 (d, JP,P = 20.9 Hz); -11.4 (d, JP,P = 21.0 Hz); m/z (ESI) 727.1276 [M-2H]-,
C25H29N8O14P2- requires 727.1284; HPLC: retention time 7.68 min (99.8%).
Chapter 7
229
7.2.6 Synthesis of 8-(pyrrol-2-yl) cADPR (24)
8-(Pyrrol-2-yl)-adenine-[P1-(9β-D-ribofuranos-5′-yl)-P
2-(1β-D-ribofuranos-5′-yl)]
pyrophosphate (24)246
A stock solution of ADPRC from Aplysia californica (50 U/mL) was prepared in
HEPES buffer (50 mM, pH 7.4). To a solution of 15h (0.005 g) in HEPES buffer (13
mL, 50 mM, pH 7.4) an appropriate volume of ADPRC stock was added to have a final
enzyme concentration of 0.192 U/mL. The reaction was incubated for 1 h at room
temperature until HPLC analysis indicated completion of the reaction. Afterwards, the
solution was evaporated to dryness and the crude product purified by purification
method 2 (0-80% 1 M TEAB buffer against H2O over 240 mL, flow rate: 2 mL/min,
fraction size: 3 mL) to give 24 as a white residue (0.0051 g, 2.76 equiv. TEA, 96%
yield). 1H NMR (400 MHz; D2O) δ: 8.98 (1H, s, 2-H), 7.23 (1H, dd, J5,3 = 1.4 Hz, 5-H
pyrrole), 6.92 (1H, dd, J3,4 = 3.0 Hz, 3-H pyrrole), 6.45 (1H, dd, J4,3 = 3.1 Hz, J4,5 = 2.9
Hz, 4-H pyrrole), 6.42 (1H, d, J = 5.3 Hz, 1′-H), 6.16 (1H, d, J = 3.3 Hz, 1′′-H), 5.53
(1H, apparent t, 2′-H), 4.50 – 4.09 (9H, m, 2′′-H, 3′-H, 3′′-H, 4′-H, 4′′-H, 5′-H2, 5′′-H2),
3.26 (16.6H, q, J = 7.3 Hz, CH2 TEA), 1.24 (24.8H, t, J = 7.3 Hz, CH3 TEA); 31
P NMR
(162 MHz; CD3OD) δ: -10.2 (d, JP,P = 11.7 Hz), -9.8 (d, JP,P = 11.3 Hz); m/z (ESI)
607.0947 [M]+, C19H25N6O13P2
+ requires 607.0949; HPLC: retention time 8.7 min.
Chapter 7
230
7.3 Quantum yield measurements for AMP and NAD+ derivatives
Fluorescence quantum yields (Фx) of AMP and NAD+ derivatives were determined by
the comparative method,204, 205
using 2-amino pyridine in 0.1 M H2SO4 (Ф = 0.60) and
tryptophan in water, pH 7.2 (Ф = 0.14), as reference standards. Absorbance spectra for
all the compounds were recorded at room temperature from the most concentrated (Abs
≤ 1) to the most diluted solution between 200 and 800 nm. Emission spectra of the same
solutions were recorded at room temperature between 200 and 900 nm at λex maxima;
the emission spectra for the same compound must be recorded with constant slit width
to make valid the obtained quantum yield values. The integrated areas under the
emission spectra were plotted as function of the integrated areas under the absorbance
spectra at corresponding concentrations. The obtained gradient, Gradx, was then
substituted in equation [1] (see page 94), where Gradst is known from the calibration of
the two standards, performed with the same protocol, and st is known from the
literature. After correction for the refractive index, η, of the solvents used in the
measurements, it is possible to calculate x from equation [1].
7.4 Chemical hydrolysis of NAD+ derivatives172
For the monitoring by fluorescence of the chemical hydrolysis of NAD+ derivatives, 0.1
N NaOH solution was added to the cuvette and the reaction was started by the addition
of a requisite volume of aqueous solution of the compounds to give a final
concentration of 0.8 μM. The basic hydrolysis was monitored at λex and λem maxima for
each compound at room temperature. In the case of 15h, the whole emission spectra at
λex maxima was recorded over time (data not shown).
The HPLC time-course for the basic hydrolysis of 15h was performed on a 49.5 μM
solution in 0.1 N NaOH at room temperature. Each sample was quenched with an
equivalent volume of 0.1 N HCl prior to injection into the HPLC. (for the HPLC
method, see 7.1.2; retention times: NaM 3.8 min, 15h 7.3 min, 23 8.0 min). Peak areas
were fitted to a 1st order rate equation using GraFit5. The estimated rate constant for the
formation of NaM, kNaM, from fitting the peak areas to a 1st order rate equation is 0.0672
min-1
, and the t1/2 = 10.31 min, in agreement within error of each other with the
Chapter 7
231
estimated rate constant for the formation of 23, k23, 0.0586 min-1
, and the t1/2 = 11.83
min.
7.5 Protocols for HPLC assays of enzyme activities172
7.5.1 Nucleotide pyrophosphatase (NPP)
Separate stock solutions of 15h (82.48 μM) and NPP from Crotalus adamanteus venom
(8.39 U/mL) were prepared in Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10
mM). The reaction was started by addition of an appropriate volume of 15h stock (300
μL, 24.74 μM) to the enzyme solution (1.2 μL, 0.01 U/mL) in the reaction buffer to give
a final volume of 1 mL (all concentrations are final concentrations). The reaction was
incubated at 30 ºC for 90 min. Samples were drawn at different time points and stored
immediately on dry-ice for 20 min, to stop the enzymatic reaction. Samples were diluted
as appropriate and analysed by HPLC (for the HPLC method, see 7.1.2; retention times:
15h 6.5 min, β-NMN 1.8 min, 9h 7.7 min). The estimated rate constant k for the
formation of 9h was obtained by fitting the peak areas to a 1st order rate equation (k9h
0.2457 min-1
, t1/2 2.82 min) using Grafit5.
7.5.2 NAD+-glycohydrolase (NADase)
A stock solution of 15h (82.48 μM) was prepared in Tris/HCl buffer (50 mM, pH 8).
NADase from porcine brain (0.01 U/mg) was accurately weighed and suspended in
Tris/HCl buffer (50 mM, pH 8) and incubated for a few minutes at 37 ºC. The reaction
was started by addition of an appropriate volume of 15h stock (300 μL, 24.74 μM) to
the enzyme suspension (0.9 mg, 0.009 U/mL) in the reaction buffer to give a final
volume of 1 mL (all concentrations are final concentrations). Samples were drawn at
different time points and stored immediately on dry-ice for 20 min, to stop the
enzymatic reaction. Samples were centrifuged, diluted as appropriate and analysed by
HPLC (for the HPLC method, see 7.1.2; retention times: 15h 7.3 min, NaM 3.8 min, 23
8.0 min). The estimated rate constants for the formation of NaM, kNaM, and the
formation of 23, k23, were obtained by fitting the peak areas to a 1st order rate equation
(kNaM 0.1750 min-1
, t1/2 3.96 min; k23 0.1696 min-1
, t1/2 = 4.09 min) using Grafit5. In a
Chapter 7
232
control experiment, 15h was incubated in Tris/HCl buffer (50 mM, pH 8) at 37 ºC over
1 h without degradation.
7.5.3 ADP-ribosyl cyclase (ADPRC)
Separate stock solutions of 15h (82.48 μM) and ADPRC from Aplysia californica (50
U/mL) were prepared in HEPES buffer (50 mM, pH 7.4). The reaction was started by
addition of an appropriate volume of 15h stock (75 μL, 12.37 μM) to the enzyme
solution (0.5 μL, 0.05 U/mL) in the reaction buffer to give a final volume of 500 μL (all
concentrations are final concentrations). The reaction was incubated at 26 ºC for 1 h.
Samples were drawn at different time points and stored immediately on dry-ice for 20
min, to stop the enzymatic reaction. Samples were diluted as appropriate and analysed
by HPLC (for the HPLC method, see 7.1.2; retention times: 15h 6.6 min, NaM 3.8 min,
24 8.7 min). The estimated rate constants for the formation of NaM, kNaM, and the
formation of 24, k24, were obtained by fitting the peak areas to a 1st order rate equation
(kNaM 0.3525 min-1
, t1/2 1.97 min; k24 0.1700 min-1
, t1/2 4.08 min) using Grafit5.
7.6 Protocols for fluorescent assays of enzyme activities172
7.6.1 Nucleotide pyrophosphatase (NPP)
Separate stock solutions of 15h (17.6 μM), 9h (9.4 μM) and NPP from Crotalus
adamanteus venom (8.39 U/mL) were prepared in Tris/HCl buffer (50 mM, pH 8)
containing MgCl2 (10 mM), unless stated otherwise. β-NAD+ (964.7 μM), AMP (1730
μM) and β-NMN (478.8 μM) stock solutions were prepared in Milli-Q water, unless
stated otherwise. In the kinetic experiments, the requisite volume of substrate 15h or
product 9h was added to individual wells of a 96-well plate, and Tris/HCl buffer (50
mM, pH 8) containing MgCl2 (10 mM) was added as required to give a final volume of
100 μL. In the inhibition experiments, the requisite volume of substrate 15h and β-
NAD+, or AMP, or β-NMN was added to individual wells of a 96-well plate, and
Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10 mM) was added as required to
give a final volume of 100 μL. In each experiment, two calibration wells were included
for the determination of sample concentration from fluorescence intensity: one well
Chapter 7
233
contained only Tris/HCl buffer (50 mM, pH 8) and MgCl2 (10 mM) and gave the „zero‟
readout (negative control), and the second contained 9h (3 μM) inTris/HCl buffer (50
mM, pH 8) and MgCl2 (10 mM) (positive control). Samples and fluorimeter readings
were allowed to stabilize by placing the plate in the microplate reader at 30 ºC and
taking readings (excitation 300 ± 5 nm, emission 410 ± 5 nm, gain 15%) until a stable
signal was obtained (typically ca 60 min). The data collection program was re-started
and, after 7 min, the enzyme reaction was initiated by addition of enzyme stock solution
(20 μL, 0.007 U/mL per well). Fluorescence emission was recorded for 60 min. All
samples, including controls, were recorded in duplicate, and the mean of the two
readings processed. To convert [RFU] to 9h concentration [μM], it was assumed that
there is very little contribution to fluorescence from 15h related to 9h. For each assay
point, the fluorescence emission [RFU], Ft, was converted to 9h concentration [μM],
[9h]t, using the equation [2] and with reference to the fluorescence from 9h (3 μM),
F3μM. From the resulting plot of [9h] vs time, initial velocity can be calculated from
linear region at the beginning of each trace.
[2]
For the determination of kinetic parameters, Km and Vmax, enzyme (0.007 U/mL) and
15h (0.2 μM – 6.5 μM) in Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10 mM)
were used (total volume per well: 120 μL, all concentrations are final concentrations).
Kinetic parameters were reported as the mean (± SD) of at least two data sets. For the
IC50 determination of β-NAD+, increasing concentrations of β-NAD
+ (0 – 350 μM) were
introduced into assays of 15h (4.4 μM) and NPP (0.007 U/mL). For the kinetic analysis
of β-NAD+ inhibition, enzyme (0.007 U/mL), 15h (0.2 μM – 5 μM) and β-NAD
+ (8 μM
– 15 μM) in Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10 mM) were used (total
volume per well: 120 μL, all concentrations are final concentrations). For the kinetic
analysis of AMP inhibition, enzyme (0.007 U/mL), 15h (0.2 μM – 5 μM) and AMP (8
μM – 40 μM) in Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10 mM) were used
(total volume per well: 120 μL, all concentrations are final concentrations). For the
kinetic analysis of β-NMN inhibition, enzyme (0.007 U/mL), 15h (0.2 μM – 5 μM) and
Chapter 7
234
β-NMN (7 μM – 17 μM) in Tris/HCl buffer (50 mM, pH 8) containing MgCl2 (10 mM)
were used (total volume per well: 120 μL, all concentrations are final concentrations).
7.6.2 NAD+-glycohydrolase (NADase)
A stock solution of 15h (6.3 μM) was prepared in Tris/HCl buffer (50 mM, pH 8),
unless stated otherwise. In the kinetic experiments, an appropriate volume of Tris/HCl
buffer (150 μL, 50 mM, pH 8) was added to individual wells of a 96-well plate.
NADase from porcine brain (0.01 U/mg) was accurately weighed and suspended in
Tris/HCl buffer (50 mM, pH 8) and incubated for several minutes at 37 ºC. The reaction
was started by addition of an appropriate volume of 15h stock (1 mL, 5.25 μM) to the
enzyme suspension (0.9 mg, 0.0075 U/mL) in the reaction buffer to give a final volume
of 1.2 mL (all concentrations are final concentrations). Samples (100 μL) were drawn at
different time points and stored immediately on dry-ice for 20 min, to stop the
enzymatic reaction. Then, from each one, 50 μL was drawn and added to the wells to
give a final volume of 200 μL. Fluorescence emission was recorded at 37 ºC (excitation
300 ± 5 nm, emission 410 ± 5 nm, gain 15%).
7.6.3 ADP-ribosyl cyclase (ADPRC)
Separate stock solutions of 15h (82.48 μM) and ADPRC (50 U/mL) from Aplysia
californica were prepared in HEPES buffer (50 mM, pH 7.4), unless stated otherwise.
In the kinetic experiments, the requisite volume of substrate 15h was added to
individual wells of a 96-well plate, and HEPES buffer (50 mM, pH 7.4) was added as
required to give a final volume of 198 μL. In each experiment, two calibration wells
were included for the determination of sample concentration from fluorescence
intensity: one well contained only HEPES buffer (50 mM, pH 7.4) and gave the „zero‟
readout (negative control), and the second contained 15h (34 μM) in HEPES buffer (50
mM, pH 7.4) (positive control). Samples and fluorimeter readings were allowed to
stabilize by placing the plate in the microplate reader at 25 ºC and taking readings
(excitation 300 ± 5 nm, emission 410 ± 5 nm, gain 15%) until a stable signal was
obtained (typically ca 40 min). The data collection program was re-started and, after 17
min, the enzyme reaction was initiated by addition of enzyme stock solution (2 μL,
Chapter 7
235
0.075 U/mL per well). Fluorescence emission was recorded for 80 min. All samples,
including controls, were recorded in duplicate, and the mean of the two readings
processed. To convert [RFU] to 15h concentration [μM], it was assumed that there is
very little contribution to fluorescence from the reaction product, 24. For each assay
point, the fluorescence emission [RFU], Ft, was converted to 15h concentration [μM],
[15h]t, using the equation [3] and with reference to the fluorescence from 15h (34 μM),
F34μM: From the resulting plot of [15h] vs time, initial velocity can be calculated from
linear region at the beginning of each trace.
[3]
For the determination of kinetic parameters, Km and Vmax, enzyme (0.075 U/mL) and
15h (2.1 μM – 33.9 μM) in HEPES buffer (50 mM, pH 7.4) were used (total volume per
well: 200 μL, all concentrations are final concentrations). Kinetic parameters were
reported as the mean (± SD) of at least two data sets.
7.7 Escherichia coli DNA ligase (EcLigA)
7.7.1 Expression and purification250
The gene encoding DNA ligase from Escherichia coli (EcLigA) was previously cloned
into pET-16b vector to introduce 10-His-tag at the N terminus of the protein and
ampicillin resistance. The resulting plasmid, pRB20, was transformed in competent E.
coli BL21 pLysS cells, plated on LB-agar containing chloramphenicol and ampicillin
antibiotics. The plates were incubated at 37 °C overnight. Single colonies were
inoculated into 5 mL LB media containing antibiotics, and incubated at 30 °C overnight.
1 mL of overnight culture was inoculated into 100 mL LB media containing antibiotics,
and incubated at 37 °C until growth had reached mid log phase (OD600 0.4 – 0.6).
Therefore, IPTG was added to 400 μM final concentration to induce the protein over-
expression, and the culture incubated at 25 °C overnight. The culture was centrifuged
and the pellets re-suspended in binding buffer (Table 7.2) and sonicated; the insoluble
fraction was separated by centrifugation. The supernatant was filtered through a 0.45
μm filter, then applied to a His-tag binding column, previously equilibrated with
Chapter 7
236
binding buffer, to purify the over-expressed protein by affinity chromatography.
Following sample loading, the column was rinsed with 10 mL binding buffer and 10 mL
wash buffer to remove eventual proteins not-specifically interacting with the column
resin; finally, the protein was eluted with 5 mL elution buffer (Table 7.2). The protein
purity was confirmed by SDS-PAGE analysis. To desalt the protein from imidazole,
size-exclusion chromatography was used. The protein concentration was determined
using the Bradford method. Glycerol (30%) and DTT (1 mM) were added to the
obtained protein, that was aliquoted and stored at -80 °C.
Buffer Composition
Binding buffer 5 mM imidazole, 0.5 M NaCl, 20 mM Tris/HCl pH 7.9
Wash buffer 60 mM imidazole, 0.5 M NaCl, 20 mM Tris/HCl pH 7.9
Elution buffer 1 M imidazole, 0.5 M NaCl, 20 mM Tris/HCl pH 7.9
Table 7.2 Composition of buffers used for nickel-affinity chromatography.
7.7.2 Preparation of DNA substrate
A 40 bp double-stranded oligonucleotide containing a single-strand nick between bases
18 and 19 was used as a substrate for the DNA ligation. It was obtained by annealing a
18-mer (LigSub1) and a 22-mer (LigSub2) to a complementary 40-mer (LigSub1+2) in
ratio 1:3:3 in 10 x TBE/KCl buffer (1.1 M Tris/borate pH 8.3, 25 mM EDTA, 1 M
KCl): in particular, the 18-mer contained a fluorescein (F) molecule in 5′-position, and
the 22-mer a phosphate group (Pi) in 5′-position (Table 7.3). The annealing was checked
by native PAGE and the substrate stored at -20 °C.
Oligonucleotide Sequence (5′ to 3′)
LigSub1 F-GTA AAA CGA CGG CCA GTG
LigSub2 Pi-AAT TCG AGC TCG GTA CCC GGG G
LigSub1+2 CCC CGG GTA CCG AGC TCG AAT TCA CTG GCC GTC GTT TTA C
Table 7.3 Oligonucleotides sequence used for the preparation of substrate for ligation assay.
Chapter 7
237
7.7.3 Ligation assay
Separate stock solutions of DNA substrate (5 μM), EcLigA (0.2 μM), β-NAD+ (181
μM) and the different compounds (1 mM) were prepared in Milli-Q water. To test
EcLigA activity, the requisite volumes of DNA substrate (0.75 μL, 0.125 μM) and β-
NAD+ (4.3 μL, 26 μM) were added to an appropriate volume of 1 x buffer (22 μL, 30
mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA) and incubated at 30 °C
under shaking: the reaction was started with EcLigA addition (3 μL, 0.1 μM) to give a
final volume of 30 μL (all concentrations are final concentrations). For the inhibition
and the NNS activity experiments of C-2 and C-6 substituted derivatives, the requisite
volumes of DNA substrate (1.5 μL, 0.5 μM), C-2 substituted derivatives (3 μL, 200
μM), or C-6 substituted derivatives (3.75 μL, 250 μM), with/without β-NAD+ (2.15 μL,
26 μM), were added to an appropriate volume of 1 x buffer (30 mM Tris/HCl pH 8, 4
mM MgCl2, 1 mM DTT, 50 μg/mL BSA) to give a final volume of 11.25 μL; the
reaction, incubated at 30 °C under shaking, was started with EcLigA addition (3.75 μL,
0.05 μM) to give a final volume of 15 μL (all concentrations are final concentrations).
For the inhibition and the NNS activity experiments of C-8 substituted derivatives, the
requisite volumes of DNA substrate (0.75 μL, 0.125 μM), C-8 substituted derivatives
(7.5 μL, 250 μM), with/without β-NAD+ (4.3 μL, 26 μM), were added to an appropriate
volume of 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL
BSA) to give a final volume of 27 μL; the reaction, incubated at 30 °C under shaking,
was started with EcLigA addition (3 μL, 0.1 μM) to give a final volume of 30 μL (all
concentrations are final concentrations). For the NNS activity experiments of 13a and
13b, the requisite volumes of DNA substrate (1.5 μL, 0.5 μM), and 13a or 13b (0.75 –
3.75 μL, 50 – 250 μM) were added to an appropriate volume of 1 x buffer (30 mM
Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL BSA) to give a final volume of
11.25 μL; the reaction, incubated at 30 °C under shaking, was started with EcLigA
addition (3.75 μL, 0.05 μM) to give a final volume of 15 μL (all concentrations are final
concentrations). In all the set of experiments, a control reaction was included to test
EcLigA activity with only β-NAD+, not containing any tested compound. Samples were
drawn at different time points and mixed with an equal volume of 1 x stop solution
(deionized formamide, 0.5 M EDTA, 1 M NaOH, 20 mg bromophenol blue), then put in
Chapter 7
238
ice. Afterwards, samples were run on a pre-warmed 15% polyacrylamide/urea
denaturing gel.
7.7.4 Gel electrophoresis
For protein analysis, SDS-PAGE with 10% SDS-polyacrylamide resolving gel and 5%
stacking gel was used. The protein samples were diluted with 4 x SDS-loading dye (200
mM Tris, 8% SDS, 0.4% bromophenol blue, 40% glycerol, 20% DME) and loaded into
the wells; a potential of 200 V was applied for 1 h in TGS buffer (25 mM Tris pH 8.3,
192 mM glycine, 0.1% (w/v) SDS buffer). The gel was visualized by staining with
Coomassie blue.
For non-denaturing gel electrophoresis of DNA, or native PAGE, a 12% polyacrylamide
gel was used. The DNA samples were diluted with 6 x loading dye (10 mM Tris/HCl
pH 7.6, 0.03% bromophenol blue, 0.03% xylene, cyanol FF, 60% glycerol, 60 mM
EDTA) and loaded into the wells; a potential of 50 V was applied for 3 h in TBE buffer
(45 mM Tris/borate pH 8.3, 1.25 mM EDTA). The gel was visualized by fluorescence
imaging using Molecular Dynamics Storm Phosphorimager.
For denaturing gel electrophoresis of DNA, a 15% polyacrylamide/urea gel was used.
The reaction samples were diluted with 1 x stop solution (20 mL deionized formamide,
400 μL 0.5 M EDTA, 40 μL 1 M NaOH, 20 mg bromophenol blue), incubated at 95 °C
for 5 min, and loaded into the wells of a pre-warmed denaturing gel; a potential of 400
V was applied for 20 min in TBE buffer (45 mM Tris/borate pH 8.3, 1.25 mM EDTA).
The gel was visualized by fluorescence imaging using Molecular Dynamics Storm
Phosphorimager.
7.7.5 Molecular modeling for DNA ligase
All calculations were performed on an Intel Core Duo 2.8 Ghz MacBook Pro. The
crystal structure of β-NAD+-bound DNA ligase from E.faecalis was obtained from the
RCSB Protein Data Bank (PDB 1TAE). The protein was prepared for molecular
modeling analysis using the MOE modeling package (CCG, Montreal, Canada). The
ligand β-NAD+ and crystallographic water molecules were removed, leaving an empty
Chapter 7
239
active center. Hydrogen atoms were added to the heavy atoms, and the residues were
assigned with the appropriate protonation states at pH 7.4. The final protein was saved
in TriposMol2 format and directly imported into the GOLD software package
(CCDC,Cambridge, UK, Version 3.2) for subsequent docking calculations. 15h was
built with the Builder module of MOE, and hydrogen atoms were assigned to all heavy
atoms in the molecule: then, it was energy minimized using the MMFF94s force field
and the conjugate gradient method, until the default derivative convergence criterion of
0.01 kcal·mol-1
·A was met. The phosphate groups were assigned a charge of −1, as
appropriate at pH 7.4. The ligand file was exported into GOLD as Tripos Mol2 files.
The ligand dockings into the catalytic site of 1TAE were performed using Gold Suite
4.12 (CCDC, Cambridge, UK). The binding site was defined by the position of β-NAD+
in the original structure and the radius was set to 15 Å. All torsion angles in 15h were
allowed to rotate freely. GoldScore was chosen as the scoring function, and the
solutions were rescored using the ChemScore scoring function. 100 docking runs were
performed for 15h. The resulting solutions were clustered on the basis of heavy atom
rmsd values (1 ). The top-ranked solutions were visually analysed using PyMol
(PyMOL Molecular Graphics System, Version 1.1, Schr dinger, LLC). To validate the
docking protocol, β-NAD+ was redocked into the empty catalytic pocket of 1TAE using
the above protocol, and the conformations of the top scored poses were compared to the
crystal structure position of the ligand. All docking poses were exported as sdf files and
imported into MOE as a database (mdb file).
7.8 Sequence alignment and 3D alignment of protein structures
A sequence alignment was undertaken for ADPRC from Aplysia californica (PDB
3ZWM, Uniprot ID P29241), archeal Sir2-Af2 (PDB 1YC2, Uniprot ID O30124) and
DNA ligase from E.faecalis (PDB 1TAE, Uniprot ID Q837V6) using the sequence
alignment tool provided by Uniprot (Universal Protein Resource,
http://www.uniprot.org).275
Pairwise 3D alignment for the chains A of the three proteins
mentioned above was undertaken using the protein comparison tools, jFATCAT-rigid
and jFATCAT-flexible, provided by RCSB Protein Data Bank (http://www.rcsb.org/).
Appendix
240
8 APPENDIX
Figure A1 HPLC time-course for the basic hydrolysis of 15h. Conditions: 15h (49.5 μM) in 0.1 N NaOH,
rt. Data from single experiment, peak area vs time [min], data plotting to a 1st order rate equation with
Grafit5.
Figure A2 a) Fluorescence calibration curves for 15h and 9h. Conditions: 15h or 9h (0.13 – 4.26 μM),
MgCl2 (10 mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all
concentrations are final concentrations. Fluorescence intensity values were fitted to a linear equation with
Excel. The estimated linear equations are, respectively: y = 2149.4x + 418.93 (r² = 0.994) for 9h, and y =
2052.9x + 62.892 (r² = 0.991) for 15h. Each data point is the mean (± SD) of experiments run in
duplicate. b) Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h: initial velocity (V0)
as function of NPP concentration. Conditions: NPP (0.0024 – 0.024 U/mL), 15h (0.09 μM), MgCl2 (10
mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm; all concentrations are final
concentrations. Initial velocity values (V0) were calculated from the reaction time-courses (insert)
measuring the fluorescence signals generated when 10% product was formed. Initial velocity values were
fitted to a polynomial equation with GraFit5.
time [min]
0 10 20 30 40
are
a H
PL
C p
ea
ks
0
40000
80000
120000
160000
200000
240000nicotinamide
23
15h
0
2000
4000
6000
8000
10000
12000
0 1 2 3 4
Flu
ore
sc
en
ce
Em
iss
ion
[R
FU
]
[9h-15h] μM
15h
9h
NPP [U/mL]
0 0.005 0.01 0.015 0.02 0.025
V [R
FU
/sec]
0
20
40
60
80
a b
-2000
8000
18000
28000
38000
48000
58000
68000
0 100 200
Flu
ore
scen
ce E
mis
sio
n
time [sec]
0.0024U/mL0.0048U/mL0.0072U/mL0.0096U/mL0.024U/mL
V0
[RFU
/se
c]
Appendix
241
Figure A3 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h: a) Hanes-Woolf and b)
Lineweaver-Burk plot. Conditions: NPP (0.007 U/mL), 15h (0.2 – 5.2 μM), MgCl2 (10 mM), Tris/HCl
buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final
concentrations. Initial velocity values (V0) were calculated from the corresponding time-courses
measuring the fluorescence signals generated when 10% product is formed. Initial velocity values were
then fitted to linear equations with Excel. The estimated linear equation for Hanes-Woolf plot is y =
651.08x + 2591.9 (r² = 0.9519); for Lineweaver-Burk plot, y = 2808.6x + 495.98 (r² = 0.9958). Each data
point is the mean (± SD) of experiments run in duplicate.
Figure A4 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15h: Michaelis-Menten plot
for β-NAD+ inhibition toward the activity of NPP from Crotalus adamanteus venom and 15h. Conditions:
NPP (0.007 U/mL), 15h (0.2 – 5 μM), β-NAD+ (8 – 15 μM), MgCl2 (10 mM), Tris/HCl buffer (50 mM,
pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final concentrations. Initial
velocity values (V0) were calculated from the corresponding time-courses measuring the fluorescence
signals generated when 10% product was formed. Each data point is the mean (± SD) of experiments run
in duplicate, data plotting with Excel.
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
0 1 2 3 4 5 6
1/V
0
1/[15h]
0
1000
2000
3000
4000
5000
6000
7000
0 1 2 3 4 5 6
[15h
]/V
0[s
ec]
[15h] mMa b
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
0 1 2 3 4 5 6
1/V
0
1/[15h]
0
1000
2000
3000
4000
5000
6000
7000
0 1 2 3 4 5 6
[15h
]/V
0[s
ec]
[15h] mMa b
-0.0002
-0.0001
0
0.0001
0.0002
0.0003
0.0004
0.0005
0.0006
0.0007
0 1 2 3 4 5 6
V0
[mM
/sec]
[15h] mM
control
[NAD] 8uM
[NAD] 12uM
[NAD] 15uM
Appendix
242
Figure A5 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15b: a) fluorescence time-
course as fluorescence emission [RFU] vs time [sec]; b) Michaelis-Menten plot. Conditions: NPP (0.007
U/mL), 15b (0.2 – 5 μM), MgCl2 (10 mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem =
410 nm, gain 15%; all concentrations are final concentrations. Initial velocity values (V0) were calculated
from the time-courses measuring the fluorescence signals generated when 10% product is formed. Initial
velocity values were fitted to the Michaelis-Menten equation with Grafit5.
Figure A6 Fluorimetric assay of NPP from Crotalus adamanteus venom and 15c: fluorescence time-
course as fluorescence emission [RFU] vs time [sec]. Conditions: NPP (0.007 U/mL), 15c (0.2 – 5 μM),
MgCl2 (10 mM), Tris/HCl buffer (50 mM, pH 8), 30 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all
concentrations are final concentrations.
[15b] mM
0 2 4 6
V [
RF
U]/
se
c
0
2
4
6
8
0
5000
10000
15000
20000
25000
30000
0 20 40 60 80
Flu
ore
scen
ce E
mis
sio
n [
RF
U]
time [sec]
[15b] 0.2uM
[15b] 1uM
[15b] 2uM
[15b] 3uM
[15b] 4uM
[15b] 5uM
neg control
a b
V0
[RFU
]/se
c
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
0 10 20 30
Flu
ore
scen
ce E
mis
sio
n [
RF
U]
time [sec]
[15c] 0.2uM
[15c] 1uM
[15c] 2uM
[15c] 3uM
[15c] 4uM
[15c] 5uM
neg control
Appendix
243
Figure A7 Fluorimetric assay of ADPRC from Aplysia californica and 15h: ADPR-cyclase activity;
fluorescence time-course as fluorescence emission [RFU] vs time [sec]. Conditions: ADPRC (0.075
U/mL), 15h (2.12 – 33.9 μM), HEPES buffer (50 mM, pH 7.4), 26 ºC, λex = 300 nm, λem = 410 nm, gain
15%. Enzyme addition at 1030 sec. Positive control: 15h (33.9 μM). Negative control: buffer. All
concentrations are final concentrations. Each data point is the mean of experiments run in duplicate.
Figure A8 Fluorimetric assay of ADPRC from Aplysia californica and 15h: Michaelis-Menten and
Hanes-Woolf plot (insert). Conditions: ADPRC (0.075 U/mL), 15h (2.12 – 33.9 μM), HEPES buffer (50
mM, pH 7.4), 26 ºC, λex = 300 nm, λem = 410 nm, gain 15%; all concentrations are final concentrations.
Initial velocity values (V0) were calculated from the corresponding reaction time-courses measuring the
fluorescence signals generated when 10% product is formed. Initial velocity values were fitted to the
Michaelis-Menten plot with Grafit5, and to the Hanes-Woolf plot (insert) with Excel (estimated linear
equation: y = 37.26x + 2947.7, r² = 0.8568). Each data point is the mean (± SD) of experiments run in
duplicate.
0
2000
4000
6000
8000
10000
12000
14000
0 1000 2000 3000
Flu
ore
scen
ce E
mis
sio
n [
RF
U]
time [sec]
neg contr
pos contr
[15h] 2.12uM
[15h] 4.24uM
[15h] 8.47uM
[15h] 12.71uM
[15h] 16.95uM
[15h] 25.42uM
[15h] 33.9uM
[15h] uM
0 50 100 150
V [uM
/se
c]
0
0.005
0.01
0.015
0.02
0
1000
2000
3000
4000
5000
0 10 20 30 40
[15h
]/V
0
[15h] mM
V0
[mM
/se
c]
Appendix
244
Figure A9 Fluorimetric assay of ADPRC from Aplysia californica and 15h: cADPR-hydrolase activity;
fluorescence time-course as fluorescence emission [RFU] vs time [sec]. Conditions: ADPRC (1.75
U/mL), 15h (2.12 – 33.9 μM), HEPES buffer (50 mM, pH 7.4), 26 ºC, λex = 300 nm, λem = 410 nm, gain
15%; all concentrations are final concentrations. Each data point is the mean of experiments run in
duplicate.
Figure A10 Comparison of hydrolytic stability for 15h and 24. Conditions: 15h and 24 respectively in
NaOH (0.1 N) and HCl (0.1 N), 26 ºC, λex = 300 nm, λem = 410 nm, gain 15%.
0
1000
2000
3000
4000
5000
6000
7000
0 5000 10000 15000
Flu
ore
sce
nce
Em
issi
on
[R
FU]
time [sec]
neg control
pos control
[15h] 2.12uM
[15h] 4.24uM
[15h] 8.48uM
[15h] 12.71uM
[15h] 16.95uM
[15h] 25.42uM
[15h] 33.9uM
0
5000
10000
15000
20000
25000
0 200 400 600 800 1000 1200 1400
Flu
ore
sce
nce
Em
issi
on
[R
FU]
time [sec]
15h in NaOH15h in HCl24 in NaOH24 in HCl
Appendix
245
Figure A11 Preparation of annealed substrate for ligation assay at different ratio
LigSub1:LigSub2:LigSub1+2. Conditions: LigSub1 at 5 μM in 10 x TBE/KCl (50 mM) buffer; analysis
by native PAGE and gel visualization by fluorescence imaging.
Figure A12 IC50 estimation for quinacrine. Conditions: DNA substrate (0.5 μM), β-NAD+ (26 μM),
EcLigA (0.05 μM), quinacrine·2HCl (0 – 5 μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1
mM DTT, 50 μg/mL BSA) incubated at 30 °C under shaking and sampled at 5 min; all concentrations are
final concentrations. Analysis of assay run on denaturing PAGE and visualized by fluorescence imaging.
Data analysed using ImageJ. Each data point is the mean (± SD) of a set of experiments run in triplicate;
data plotting with Excel.
-5
0
5
10
15
20
25
30
35
40
0 1 2 3 4 5 6
% li
gati
on
[quinacrine] mM
Appendix
246
Figure A13 Denaturing PAGE for the inhibition of 2-substituted AMP/AMP-morpholidate/NAD+
derivatives at 200 μM. Conditions: DNA substrate (0.5 μM), β-NAD+ (26 μM), EcLigA (0.05 μM), 2-
substituted derivative (200 μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50
μg/mL BSA) incubated at 30 °C under shaking and sampled at 5 (a) and 15 min (b); all concentrations are
final concentrations. Control reaction at the same conditions without any 2-substituted derivative. Gel
visualization by fluorescence imaging.
Figure A14 IC50 estimation for 7a (a) and 13a (b). Conditions: DNA substrate (0.5 μM), β-NAD+ (26
μM), EcLigA (0.05 μM), 7a or 13a (0 – 250 μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1
mM DTT, 50 μg/mL BSA) incubated at 30 °C under shaking and sampled at 5 min; all concentrations are
final concentrations. Analysis of assays run by denaturing PAGE; visualization by fluorescence imaging
and data analysis by ImageJ. Each data point is the mean (± SD) of a set of experiments run in triplicate;
data plotting with Excel.
a b
0
1
2
3
4
5
6
7
8
9
10
0 50 100 150 200 250 300
% li
gati
on
[7a] mM
0
5
10
15
20
25
30
35
40
45
0 50 100 150 200 250 300
% li
gati
on
[13a] mMa b
Appendix
247
Figure A15 IC50 estimation for 11a. Conditions: DNA substrate (0.5 μM), β-NAD+ (26 μM), EcLigA
(0.05 μM), 11a (0 – 250 μM) in 1 x buffer (30 mM Tris/HCl pH 8, 4 mM MgCl2, 1 mM DTT, 50 μg/mL
BSA) incubated at 30 °C under shaking and sampled at 5 min. Analysis of assays run by denaturing
PAGE; visualization by fluorescence imaging and data analysis by ImageJ. Each data point is the mean (±
SD) of a set of experiments run in triplicate; data plotting with Excel.
Enzyme (PBD) Twists P RMSD Score % Id % Similarity
1TAE (chain A) 0 0.454 8.96 108.39 2.5 13.57
1YC2 (chain A) 5 0.199 3.98 163.83 4.1 12.03
1TAE (chain A) 0 0.460 9.58 100.28 2.08 7.72
3ZWM (chain A) 5 0.525 4.98 131.42 4.59 10.09
1YC2 (chain A) 0 0.181 11.12 135.22 8.27 17.29
3ZWM (chain A) 4 0.0916 18.64 150.27 4.17 12.50
Table A1 Structure pairwise alignment of ADPRC from Aplysia californica (PDB 3ZWM), archaeal Sir2-
Af2 (PDB 1YC2), and DNA ligase from E. faecalis (PDB 1TAE) by rigid FATCAT and flexible
FATCAT. The flexible structure alignment works by identifying aligned fragment pairs and allowing
flexibility in connecting them by twists in the structure. The more twists that are introduced in the
structure, the less is the similarity between the two proteins. Twists = 0 in rigid FATCAT; twists ≥ 0 in
flexible FATCAT. P is the probability value that two structures are similar (values < 0.05 indicates
significant similarities); RMSD is the average distance between the Cα atoms of superimposed proteins;
Score is the raw score of the alignment; % Id is the percent sequence identity in the alignment; %
Similarity is the percent structure similarity in the alignment.
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200 250 300
% li
gati
on
[11a] mM
References
248
9 REFERENCES
1. D. Voet, C. W. Pratt and J. G. Voet, Fundamentals of Biochemistry, 2008.
2. R. F. Anderson, Biochim. Biophys. Acta - Bioenergetics, 1980, 590, 277-281.
3. J. A. Farrington, E. J. Land and A. J. Swallow, Biochim. Biophys. Acta -
Bioenergetics, 1980, 590, 273-276.
4. M. Ziegler, Eur. J. Biochem., 2000, 267, 1550-1564.
5. P. Belenky, K. L. Bogan and C. Brenner, Trends Biochem. Sci., 2007, 32, 12-19.
6. A. Legutko, P. Lekeux and F. Bureau, Open Immunol. J., 2009, 2, 42-51.
7. H. Lin, Org. Biomol. Chem., 2007, 5, 2541-2554.
8. N. J. Oppenheimer, Mol. Cell. Biochem., 1994, 138, 245-251.
9. R. W. Johnson, T. M. Marschner and N. J. Oppenheimer, J. Am. Chem. Soc.,
1988, 110, 2257-2263.
10. C. C. Guilbert and S. L. Johnson, Biochemistry, 1977, 16, 335-344.
11. S. L. Johnson and D. L. Morrison, J. Biol. Chem., 1970, 245, 4519-4524.
12. S.-J. Lin and L. Guarente, Curr. Opin. Cell Biol., 2003, 15, 241-246.
13. N. Pollak, C. Dolle and M. Ziegler, Biochem. J., 2007, 402, 205-218.
14. J. Du, H. Jiang and H. Lin, Biochemistry, 2009, 48, 2878-2890.
15. S. Shuman, J. Biol. Chem., 2009, 284, 17365-17369.
16. M. Bollen, R. Gijsbers, H. Ceulemans, W. Stalmans and C. Stefan, Crit. Rev.
Biochem. Mol. Biol., 2000, 35, 393-432.
17. G. G. Yegutkin, Biochim. Biophys. Acta - Mol. Cell Res., 2008, 1783, 673-694.
18. G. Magni, A. Amici, M. Emanuelli, G. Orsomando, N. Raffaelli and S. Ruggieri,
Cell. Mol. Life Sci., 2004, 61, 19-34.
19. J. W. Rowen and A. Kornberg, J. Biol. Chem., 1951, 193, 497-507.
20. B. D. Bennett, E. H. Kimball, M. Gao, R. Osterhout, S. J. Van Dien and J. D.
Rabinowitz, Nat. Chem. Biol., 2009, 5, 593-599.
21. J. A. Khan, F. Forouhar, X. Tao and L. Tong, Expert Opin. Ther. Targets, 2007,
11, 695-705.
22. A. Mattevi, Nat. Struct. Mol. Biol., 2006, 13, 563-564.
References
249
23. J. R. Revollo, A. A. Grimm and S.-i. Imai, J. Biol. Chem., 2004, 279, 50754-
50763.
24. L. J. Zatman, N. O. Kaplan and S. P. Colowick, J. Biol. Chem., 1953, 200, 197-
212.
25. J. C. Tanny and D. Moazed, Proc. Natl. Acad. Sci. USA, 2001, 98, 415-420.
26. M. Kotaka, R. Graeff, Z. Chen, L. H. Zhang, H. C. Lee and Q. Hao, J. Mol.
Biol., 2012, 415, 514-526.
27. G. T. Marsischky, B. A. Wilson and R. J. Collier, J. Biol. Chem., 1995, 270,
3247-3254.
28. D. Slade, M. S. Dunstan, E. Barkauskaite, R. Weston, P. Lafite, N. Dixon, M.
Ahel, D. Leys and I. Ahel, Nature, 2011, 477, 616-620.
29. A. Ruf, V. Rolli, G. de Murcia and G. E. Schulz, J. Mol. Biol., 1998, 278, 57-65.
30. Q. Liu, I. A. Kriksunov, H. Jiang, R. Graeff, H. Lin, H. C. Lee and Q. Hao,
Chem. Biol., 2008, 15, 1068-1078.
31. K. Zhao, R. Harshaw, X. Chai and R. Marmorstein, Proc. Natl. Acad. Sci. USA,
2004, 101, 8563-8568.
32. B. C. Smith and J. M. Denu, J. Am. Chem. Soc., 2007, 129, 5802-5803.
33. L. Franco, L. Guida, S. Bruzzone, E. Zocchi, C. Usai and A. D. Flora, FASEB J.,
1998, 12, 1507-1520.
34. H. Maruta, N. Matsumura and S.-i. Tanuma, Biochem. Biophys. Res. Comm.,
1997, 236, 265-269.
35. E. N. Chini, Curr. Pharm. Des., 2009, 15, 57-63.
36. M. Kruszewski and I. Szumiel, DNA Repair, 2005, 4, 1306-1313.
37. A. Grahnert, A. Grahnert, C. Klein, E. Schilling, J. Wehrhahn and S. Hauschildt,
Innate Immun., 2011, 17, 212-233.
38. F. Malavasi, S. Deaglio, G. Zaccarello, A. L. Horenstein, A. Chillemi, V.
Audrito, S. Serra, M. Gandione and A. a. T. Zitella, A, J. Mol. Endocrinol.,
2010, 45, 183-191.
39. T. Vaisitti, V. Audrito, S. Serra, C. Bologna, D. Brusa, F. Malavasi and S.
Deaglio, FEBS Lett., 2011, 585, 1514-1520.
40. S. Bruzzone, L. Guida, E. Zocchi, L. Franco and A. De Flora, FASEB J., 2001,
15, 10-12.
41. A. Gerth, K. Nieber, N. J. Oppenheimer and S. Hauschildt, Biochem. J., 2004,
382, 849-856.
References
250
42. L. Guida, S. Bruzzone, L. Sturla, L. Franco, E. Zocchi and A. De Flora, J. Biol.
Chem., 2002, 277, 47097-47105.
43. I. Lange, S. Yamamoto, S. Partida-Sanchez, Y. Mori, A. Fleig and R. Penner,
Sci. Signal., 2009, 2.
44. R. A. Billington, E. A. Bellomo, E. M. Floriddia, J. Erriquez, C. Distasi and A.
A. Genazzani, FASEB J., 2006, 20, 521-523.
45. S. Todisco, G. Agrimi, A. Castegna and F. Palmieri, J. Biol. Chem., 2006, 281,
1524-1531.
46. A. Leskovar and J. Reinstein, Arch. Biochem. Biophys., 2008, 473, 16-24.
47. D. M. Jameson and J. F. Eccleston, in Methods Enzymol., ed. M. L. J. Ludwig
Brand, Academic Press, 1997, pp. 363-390.
48. S. E. Pollack, T. Uchida and D. S. Auld, J. Protein Chem., 1983, 2, 1-12.
49. J. Zhang, in Methods Enzymol., ed. J. W. S. C. W. Donald B. McCormick,
Academic Press, 1997, pp. 255-265.
50. K. W. Pankiewicz, Pharmacol. Therapeut., 1997, 76, 89-100.
51. P. Franchetti, L. Cappellacci, P. Perlini, H. N. Jayaram, A. Butler, B. P.
Schneider, F. R. Collart, E. Huberman and M. Grifantini, J. Med. Chem., 1998,
41, 1702-1707.
52. L. Bonnac, L. Chen, R. Pathak, G. Gao, Q. Ming, E. Bennett, K. Felczak, M.
Kullberg, S. E. Patterson, F. Mazzola, G. Magni and K. W. Pankiewicz, Bioorg.
Med. Chem. Lett., 2007, 17, 1512-1515.
53. V. Cepeda, M. A. Fuertes, J. Castilla, C. Alonso, C. Quevedo, M. Soto and J. M.
Perez, Recent Pat. Anti-Cancer Drug Discov., 2006, 1, 39-53.
54. R. J. Griffin, N. J. Curtin, D. R. Newell, B. T. Golding, B. W. Durkacz and A.
H. Calvert, Biochimie, 1995, 77, 408-422.
55. J. L. Avalos, K. M. Bever and C. Wolberger, Mol. Cell, 2005, 17, 855-868.
56. K. Inageda, K. Takahashi, K.-i. Tokita, H. Nishina, Y. Kanaho, I. Kukimoto, K.
Kontani, S.-i. Hoshino and T. Katadat, J. Biochem., 1995, 117, 125-131.
57. M. T. Schmidt, B. C. Smith, M. D. Jackson and J. M. Denu, J. Biol. Chem.,
2004, 279, 40122-40129.
58. M. E. Migaud, R. L. Pederick, V. C. Bailey and B. V. L. Potter, Biochemistry,
1999, 38, 9105-9114.
59. J. T. Slama and A. M. Simmons, Biochemistry, 1988, 27, 183-193.
60. K. A. Wall, M. Klis, J. Kornet, D. Coyle, J. C. Amé, M. K. Jacobson and J. T.
Slama, Biochem. J., 1998, 335, 631-636.
References
251
61. A. A. Sauve and V. L. Schramm, Biochemistry, 2002, 41, 8455-8463.
62. H. M. Muller-Steffner, O. Malver, L. Hosie, N. J. Oppenheimer and F. Schuber,
J. Biol. Chem., 1992, 267, 9606-9611.
63. Q. Liu, R. Graeff, I. A. Kriksunov, H. Jiang, B. Zhang, N. Oppenheimer, H. Lin,
B. V. L. Potter, H. C. Lee and Q. Hao, J. Biol. Chem., 2009, 284, 27637-27645.
64. P. E. Smith, J. Am. Chem. Soc., 1999, 121, 8637-8644.
65. J. K. Crane, I. Shulgina and T. M. Naeher, Purinergic Signal., 2007, 3, 233-246.
66. K. W. Pankiewicz, K. Lesiak and K. A. Watanabe, J. Am. Chem. Soc., 1997, 28,
3691-3695.
67. K. Felczak, L. Chen, D. Wilson, J. Williams, R. Vince, R. Petrelli, H. N.
Jayaram, P. Kusumanchi, M. Kumar and K. W. Pankiewicz, Bioorg. Med.
Chem., 2011, 19, 1594-1605.
68. R. Petrelli, Y. Y. Sham, L. Chen, K. Felczak, E. Bennett, D. Wilson, C. Aldrich,
J. S. Yu, L. Cappellacci, P. Franchetti, M. Grifantini, F. Mazzola, M. Di Stefano,
G. Magni and K. W. Pankiewicz, Bioorg. Med. Chem., 2009, 17, 5656-5664.
69. R. M. Graeff, T. F. Walseth, H. K. Hill and H. C. Lee, Biochemistry, 1996, 35,
379-386.
70. T. Pesnot, J. Kempter, J. r. Schemies, G. Pergolizzi, U. Uciechowska, T. Rumpf,
W. Sippl, M. Jung and G. K. Wagner, J. Med. Chem., 2011, 54, 3492-3499.
71. B. Zhang, H. Muller-Steffner, F. Schuber and B. V. L. Potter, Biochemistry,
2007, 46, 4100-4109.
72. J. Trapp, A. Jochum, R. Meier, L. Saunders, B. Marshall, C. Kunick, E. Verdin,
P. Goekjian, W. Sippl and M. Jung, J. Med. Chem., 2006, 49, 7307-7316.
73. S. D. Mills, A. E. Eakin, E. T. Buurman, J. V. Newman, N. Gao, H. Huynh, K.
D. Johnson, S. Lahiri, A. B. Shapiro, G. K. Walkup, W. Yang and S. S. Stokes,
Antimicrob. Agents Chemother., 2011, 55, 1088-1096.
74. S. S. Stokes, H. Huynh, M. Gowravaram, R. Albert, M. Cavero-Tomas, B. Chen,
J. Harang, J. T. Loch Iii, M. Lu, G. B. Mullen, S. Zhao, C.-F. Liu and S. D.
Mills, Bioorg. Med. Chem. Lett., 2011, 21, 4556-4560.
75. J. Hendle, A. F. Buckmann, W. Aehle, D. Schomburg and R. D. Schmid, Eur. J.
Biochem., 1993, 213, 947-956.
76. G. W. E. Plaut, C. P. Cheung, R. J. Suhadolnik and T. Aogaichi, Biochemistry,
1979, 18, 3430-3438.
77. N. O. Kaplan, M. M. Ciotti and F. E. Stolzenbach, J. Biol. Chem., 1956, 221,
833-844.
References
252
78. B. M. Anderson and N. O. Kaplan, J. Biol. Chem., 1959, 234, 1226-1232.
79. S. L. Oei, J. Griesenbeck, G. Buchlow, D. Jorcke, P. Mayer-Kuckuk, T. Wons
and M. Ziegler, FEBS Lett., 1996, 397, 17-21.
80. H. Jiang, J. H. Kim, K. M. Frizzell, W. L. Kraus and H. Lin, J. Am. Chem. Soc.,
2010, 132, 9363-9372.
81. H. Jiang, J. Congleton, Q. Liu, P. Merchant, F. Malavasi, H. C. Lee, Q. Hao, A.
Yen and H. Lin, J. Am. Chem. Soc., 2009, 131, 1658-1659.
82. A. C. Nottbohm, R. S. Dothager, K. S. Putt, M. T. Hoyt and P. J. Hergenrother,
Angew. Chem. Int. Ed., 2007, 46, 2066-2069.
83. N. J. Leonard, Crit. Rev. Biochem., 1984, 15, 125-199.
84. S. P. Colowick, N. O. Kaplan and M. M. Ciotti, J. Biol. Chem., 1951, 191, 447-
459.
85. W. Lohmann, C. Lohmann and M. Ibrahim, Naturwissenschaften, 1988, 75,
141-142.
86. A. A. Heikal, Biomark. Med., 2010, 4, 241-263.
87. J. R. Barrio, J. A. Secrist and N. J. Leonard, Proc. Natl. Acad. Sci. USA, 1972,
69, 2039–2042.
88. B. A. Gruber and N. J. Leonard, Proc. Natl. Acad. Sci. USA, 1975, 72, 3966-
3969.
89. C. D. Muller, C. Tarnus and F. Schuber, Biochem. J., 1984, 223, 715-721.
90. Z. Fang-Jie and C. J. Sih, Bioorg. Med. Chem. Lett., 1995, 5, 1701-1706.
91. B. M. Anderson and D. A. Yost, in Methods Enzymol., ed. D. B. M. Frank
Chytil, Academic Press, 1986, pp. 169-173.
92. B. M. Klebl and D. Pette, Anal. Biochem., 1996, 239, 145-152.
93. S. Armstrong and A. R. Merrill, Anal. Biochem., 2001, 292, 26-33.
94. K. Schweitzer, G. W. Mayr and A. H. Guse, Anal. Biochem., 2001, 299, 218-
226.
95. R. E. Davis, V. Mysore, J. C. Browning, J. C. Hsieh, Q.-A. T. Lu and P. D.
Katsikis, J. Histochem. Cytochem., 1998, 46, 1279-1289.
96. M. Shirato, S. Tozawa, D. Maeda, M. Watanabe, H. Nakagama and M.
Masutani, Biochem. Biophys. Res. Comm., 2007, 355, 451-456.
97. J. L. Sims, H. Juarez-Salinas and M. K. Jacobson, Anal. Biochem., 1980, 106,
296-306.
References
253
98. K. Ikai, K. Ueda and O. Hayaishi, J. Histochem. Cytochem., 1980, 28, 670-676.
99. K. S. Putt and P. J. Hergenrother, Anal. Biochem., 2004, 326, 78-86.
100. Y. Feng, J. Wu, L. Chen, C. Luo, X. Shen, K. Chen, H. Jiang and D. Liu, Anal.
Biochem., 2009, 395, 205-210.
101. H. Nakamura and Z. Tamura, Anal. Chem., 1978, 50, 2047-2051.
102. D. Wegener, F. Wirsching, D. Riester and A. Schwienhorst, Chem. Biol., 2003,
10, 61-68.
103. D. Wegener, C. Hildmann, D. Riester and A. Schwienhorst, Anal. Biochem.,
2003, 321, 202-208.
104. B. Heltweg, F. Dequiedt, E. Verdin and M. Jung, Anal. Biochem., 2003, 319, 42-
48.
105. B. O. S. Scott, M. Lavesa-Curto, D. R. Bullard, J. N. Butt and R. P. Bowater,
Anal. Biochem., 2006, 358, 90-98.
106. L. Liu, Z. Tang, K. Wang, W. Tan, J. Li, Q. Guo, X. Meng and C. Ma, Analyst,
2005, 130, 350-357.
107. C.-Y. Lee, D. A. Lappi, B. Wermuth, J. Everse and N. O. Kaplan, Arch.
Biochem. Biophys., 1974, 163, 561-569.
108. J. L. Panza, A. J. Russell and E. J. Beckman, Tetrahedron, 2002, 58, 4091-4104.
109. R. Parkesh, A. M. Lewis, P. K. Aley, A. Arredouani, S. Rossi, R. Tavares, S. R.
Vasudevan, D. Rosen, A. Galione, J. Dowden and G. C. Churchill, Cell
Calcium, 2008, 43, 531-538.
110. J. Lee, H. Churchil, W.-B. Choi, J. E. Lynch, F. E. Roberts, R. P. Volante and P.
J. Reider, Chem. Commun., 1999, 729-730.
111. N. E. Batoux, F. Paradisi, P. C. Engel and M. E. Migaud, Tetrahedron, 2004, 60,
6609-6617.
112. B. Zhang, G. K. Wagner, K. Weber, C. Garnham, A. J. Morgan, A. Galione, A.
H. Guse and B. V. L. Potter, J. Med. Chem., 2008, 51, 1623-1636.
113. H. Gold, P. van Delft, N. Meeuwenoord, J. D. C. Cod e, D. V. Filippov, G.
Eggink, H. S. Overkleeft and G. A. van der Marel, J. Org. Chem., 2008, 73,
9458-9460.
114. L. A. Agrofoglio, I. Gillaizeau and Y. Saito, Chem. Rev., 2003, 103, 1875-1916.
115. H. Rosemeyer, Chem. Biodivers., 2004, 1, 361-401.
116. M. Legraverend, Tetrahedron, 2008, 64, 8585-8603.
117. J. H. Lister, The chemistry of heterocyclic compounds, Wiley Interscience, 1971.
References
254
118. E. Y. Sutcliffe and R. K. Robins, J. Org. Chem., 1963, 28, 1662-1666.
119. B. C. Pal, Biochemistry, 1962, 1, 558-563.
120. M. J. J. Jacob Strouse, and Jeffrey B. Arterburn, Acta Chim. Slov., 2005, 52,
187–199.
121. M. Hocek, Eur. J. Org. Chem., 2003, 2, 245-254.
122. J. Liu and M. J. Robins, Org. Lett., 2005, 7, 1149-1151.
123. M. Hocek, Nucleic Acids Symp. Ser., 2005, 49, 29-30.
124. M. K. Lakshman, J. Organomet. Chem., 2002, 653, 234-251.
125. M. K. Lakshman, Curr. Org. Synth., 2005, 2, 83-112.
126. P. Gunda, L. M. Russon and M. K. Lakshman, Angew. Chem. Int. Ed., 2004, 43,
6372-6377.
127. M. K. Lakshman, J. H. Hilmer, J. Q. Martin, J. C. Keeler, Y. Q. V. Dinh, F. N.
Ngassa and L. M. Russon, J. Am. Chem. Soc., 2001, 123, 7779-7787.
128. I. Era, R. Pohl, B. Klepetov and M. Hocek, Org. Lett., 2006, 8, 5389-5392.
129. I. Cerna, R. Pohl and M. Hocek, Chem. Commun., 2007, 4729-4730.
130. T. E. Storr, J. A. Strohmeier, C. G. Baumann and I. J. S. Fairlamb, Chem.
Commun., 2010, 46, 6470-6472.
131. N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457-2483.
132. J. K. Stille, Angew. Chem. Int. Ed., 1986, 25, 508-524.
133. S. Kotha, K. Lahiri and D. Kashinath, Tetrahedron, 2002, 58, 9633–9695.
134. A. Matsuda, M. Shinozaki, T. Yamaguchi, H. Homma, R. Nomoto, T. Miyasaka,
Y. Watanabe and T. Abiru, J. Med. Chem., 1992, 35, 241-252.
135. H. Homma, Y. Watanabe, T. Abiru, T. Murayama, Y. Nomura and A. Matsuda,
J. Med. Chem., 1992, 35, 2881-2890.
136. L. Chen, G. Gao, K. Felczak, L. Bonnac, S. E. Patterson, D. Wilson, E. M.
Bennett, H. N. Jayaram and L. Hedstrom, J. Med. Chem., 2007, 50, 5743-5751.
137. T. Persson, S. Gronowitz, A.-B. Hörnfeldt and N. G. Johansson, Bioorg. Med.
Chem., 1995, 3, 1377-1382.
138. J. o. Neres, N. P. Labello, R. V. Somu, H. I. Boshoff, D. J. Wilson, J. Vannada,
L. Chen, C. E. Barry, E. M. Bennett and C. C. Aldrich, J. Med. Chem., 2008, 51,
5349-5370.
139. V. Nair and D. A. Young, J. Org. Chem., 1985, 50, 406-408.
References
255
140. M. J. Robins and B. Uznanski, Can. J. Chem., 1981, 59, 2601-2607.
141. Y. Xu, H. Jin, Z. Yang, L. Zhang and L. Zhang, Tetrahedron, 2009, 65, 5228-
5239.
142. C. N. Johnson, G. Stemp, N. Anand, S. C. Stephen and T. Gallagher, Synlett,
1998, 1025-1027.
143. K. L. Billingsley, K. W. Anderson and Buchwald, Angew. Chem. Int. Ed., 2006,
45, 3484-3488.
144. A. Collier and G. Wagner, Org. Biomol. Chem., 2006, 4, 4526-4532.
145. M. Yoshikawa, T. Kato and T. Takenishi, Bull. Chem. Soc. Jpn, 1969, 42, 3505-
3508.
146. T. Sowa and S. Ouchi, Bull. Chem. Soc. Jpn, 1975, 48, 2084-2090.
147. M. Hocek, A. Holý, I. Votruba and H. Dvořáková, J. Med. Chem., 2000, 43,
1817-1825.
148. M. Hocek, P. Nauš, R. Pohl, I. Votruba, P. A. Furman, P. M. Tharnish and M. J.
Otto, J. Med. Chem., 2005, 48, 5869-5873.
149. P. Nauš, R. Pohl, I. Votruba, P. D ubák, M. n. Hajd ch, R. Ameral, G. Birkuš,
T. Wang, A. S. Ray, R. Mackman, T. Cihlar and M. Hocek, J. Med. Chem.,
2009, 53, 460-470.
150. P. Capek, R. Pohl and M. Hocek, Org. Biomol. Chem., 2006, 4, 2278-2284.
151. A. Collier and G. K. Wagner, Chem. Commun., 2008, 2, 178-180.
152. T. Pesnot and G. K. Wagner, Org. Biomol. Chem., 2008, 6, 2884-2891.
153. I. Hasan, E. R. Marinelli, L.-C. C. Lin, F. W. Fowler and A. B. Levy, J. Org.
Chem., 1981, 46, 157-164.
154. S. R. Dandepally and A. L. Williams, Tetrahedron Lett., 2009, 50, 1071-1074.
155. M. Ikehara, S. Uesugi and M. Kaneko, Chem. Commun. (London), 1967, 1, 17-
18.
156. K. Mikhailov, N. Marchenkov, V. Chichikina, V. Orlova and N. Myasoedov,
Chem. Nat. Compd., 1976, 12, 461-464.
157. E. C. Western, J. R. Daft, E. M. Johnson, P. M. Gannett and K. H. Shaughnessy,
J. Org. Chem., 2003, 68, 6767-6774.
158. A. Collier and G. K. Wagner, ChemInform, 2007, 38, 3713-3721.
159. K. H. Shaughnessy, Chem. Rev., 2009, 109, 643-710.
160. N. Kohyama, T. Katashima and Y. Yamamoto, Synthesis, 2004, 2799-2804.
References
256
161. N. Kudo and M. Perseghini, Angew. Chem. Int. Ed., 2006, 45, 1282–1284.
162. C. A. Fleckenstein and H. Plenio, Chemistry, 2008, 14, 4267–4279.
163. D. G. Hall, Structure, Properties, and Preparation of Boronic Acid Derivatives:
Overview of Their Reactions and Applications., Wiley-VCH Verlag GmbH &
Co KGaA, Weinheim, Germany, Boronic Acids, 2005.
164. R. B. Bedford, M. E. Blake, C. P. Butts and D. Holder, Chem. Commun., 2003,
4, 466-467.
165. B. H. Lipshutz and A. R. Abela, Org. Lett., 2008, 10, 5329-5332.
166. B. H. Lipshutz, T. B. Petersen and A. R. Abela, Org. Lett., 2008, 10, 1333-1336.
167. V. A. Bloomfield, D. M. Crothers and I. Tinoco, Nucleic Acids: Structures,
Properties, and Functions, Sausalito, Calif, University Science Books, 2000.
168. J. G. Moffatt and H. G. Khorana, J. Am. Chem. Soc., 1961, 83, 663-675.
169. S. Roseman, J. J. Distler, J. G. Moffatt and H. G. Khorana, J. Am. Chem. Soc.,
1961, 83, 659-663.
170. T. Mukaiyama and M. Hashimoto, Bull. Chem. Soc. Jpn, 1971, 44, 2284.
171. M. Hashimoto, M. Ueki and T. Mukaiyama, Chem. Lett., 1976, 5, 157-160.
172. G. Pergolizzi, J. N. Butt, R. P. Bowater and G. K. Wagner, Chem. Commun.,
2011, 47, 12655-12657.
173. S. L. Beaucage and M. H. Caruthers, Tetrahedron Lett., 1981, 22, 1859-1862.
174. R. L. Letsinger and W. B. Lunsford, J. Am. Chem. Soc., 1976, 98, 3655–3661.
175. M. D. Matteucci and M. H. Caruther, J. Am. Chem. Soc., 1981, 103, 3185-3191.
176. M. L. Paul, D. L. Miles and M. A. Cook, J. Pharm. Exp. Ther., 1982, 222, 241-
245.
177. C. D. Jardetzky, J. Am. Chem. Soc., 1960, 82, 229-233.
178. C. F. G. C. Geraldes and H. Santos, Perkin Trans. 2, 1983, 11, 1693-1697.
179. R. Stolarski, L. Dudycz and D. Shugar, Eur. J. Biochem., 1980, 108, 111-121.
180. C. Altona and M. Sundaralingam, J. Am. Chem. Soc., 1973, 95, 2333-2244.
181. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, 2006.
182. E. L. Wehry, in Handbook of Instrumental Techniques for Analytical Chemistry,
ed. F. A. Settle, Prentice Hall PTR, 1997.
183. M. Sauer, J. Hofkens and J. Enderlein, Handbook of Fluorescence Spectroscopy
and Imaging, Wiley VCH, 2011.
References
257
184. R. W. Sinkeldam, N. J. Greco and Y. Tor, Chem. Rev., 2010, 110, 2579-2619.
185. L. J. Kricka and P. Fortina, Clin. Chem., 2009, 55, 670-683.
186. J. W. Longworth, R. O. Rahn and R. G. Shulman, J. Chem. Phys., 1966, 45,
2930-2939.
187. E. Nir, K. Kleinermanns, L. Grace and M. S. de Vries, J. Phys. Chem. A, 2001,
105, 5106-5110.
188. J.-M. L. Pecourt, J. Peon and B. Kohler, J. Am. Chem. Soc., 2001, 123, 10370-
10378.
189. A. M. Ababneh, C. C. Large and S. Georghiou, J. Appl. Sci. Res., 2007, 3, 1288-
1295.
190. J. N. Wilson and E. T. Kool, Org. Biomol. Chem., 2006, 4, 4265-4274.
191. J. N. Wilson, J. Gao and E. T. Kool, Tetrahedron, 2007, 63, 3427-3433.
192. S. Paszyc and M. Rafalska, Nucleic Acids Res., 1979, 6, 385–397.
193. D. S. S. Zhou, S. C. Pomerantz, P. F. Crain, J. A. McCloskey, Nucleosides
Nucleotides Nucleic Acids, 2004, 23, 41-50.
194. L. M. Wilhelmsson, P. Sandin, A. Holmén, B. Albinsson, P. Lincoln and B.
Nordén, J. Phys. Chem. B, 2003, 107, 9094-9101.
195. J. C. Greenfield, N. J. Leonard and R. I. Gumport, Biochemistry, 1975, 14, 698-
706.
196. E. Sharon, S. A. Lévesque, M. N. Munkonda, J. Sévigny, D. Ecke, G. Reiser and
B. Fischer, ChemBioChem, 2006, 7, 1361-1374.
197. Y. Shinohara, K. Matsumoto, K. Kugenuma, T. Morii, Y. Saito and I. Saito,
Bioorg. Med. Chem. Lett., 2010, 20, 2817-2820.
198. A. G. Firth, I. J. S. Fairlamb, K. Darley and C. G. Baumann, Tetrahedron Lett.,
2006, 47, 3529-3533.
199. D. W. Dodd, K. N. Swanick, J. T. Price, A. L. Brazeau, M. J. Ferguson, N. D.
Jones and R. H. E. Hudson, Org. Biomol. Chem., 2010, 8, 663-666.
200. L. Zilbershtein, A. Silberman and B. Fischer, Org. Biomol. Chem., 2011, 9,
7763-7773.
201. S. G. Srivatsan, H. Weizman and Y. Tor, Org. Biomol. Chem., 2008, 6, 1334-
1338.
202. Y. T. Nicholas J. Greco, Tetrahedron, 2007, 63, 3515-3527.
203. T. M. Michiko Kimoto, Yoko Harada, Akira Sato, Shigeyuki Yokoyama, and
Ichiro Hirao, Nucleic Acids Res., 2007, 35, 5360–5369.
References
258
204. A. T. R. Williams, S. A. Winfield and J. N. Miller, Analyst, 1983, 108, 1067-
1071.
205. Jobin Yvon Ltd,
http://www.horiba.com/fileadmin/uploads/Scientific/Documents/Fluorescence/q
uantumyieldstrad.pdf, 1996.
206. P. I. Vestues and R. B. Martin, J. Am. Chem. Soc., 1981, 103, 806-809.
207. A. J. Lees, Chem. Rev., 1987, 87, 711-743.
208. C. E. Moore Jr and A. L. Underwood, Anal. Biochem., 1969, 29, 149-153.
209. P. E. Smith and J. J. Tanner, J. Mol. Recogn., 2000, 13, 27-34.
210. J. E. Roberts, Y. Aizono, M. Sonenberg and N. I. Swislocki, Bioorg. Chem.,
1975, 4, 181-187.
211. D. G. Cross and H. F. Fisher, Biochemistry, 1969, 8, 1147-1155.
212. A. Schmidt, A. Lindner, M. Nieger, M. Ruiz-Delgado and F. J. Ramirez, Org.
Biomol. Chem., 2006, 4, 3056-3066.
213. P. R. Mitchell and H. Sigel, Eur. J. Biochem., 1978, 88, 149-154.
214. R. H. Sarma and R. J. Mynott, J. Am. Chem. Soc., 1973, 95, 7470-7480.
215. O. Jardetzky and N. G. Wade-Jardetzky, J. Biol. Chem., 1966, 241, 85-91.
216. G. Zieger, R. Konrat and H. Sterk, Magn. Res. Chem., 1992, 30, S3-S6.
217. R. M. Riddle, T. J. Williams, T. A. Bryson, R. B. Dunlap, R. R. Fisher and P. D.
Ellis, J. Am. Chem. Soc., 1976, 98, 4286-4290.
218. C.-H. Lee and I. Tinoco, Biochemistry, 1977, 16, 5403-5414.
219. C. Stefan, S. Jansen and M. Bollen, Purinergic Signal., 2006, 2, 361–370.
220. J. W. Goding, B. Grobben and H. Slegers, Biochim. Biophys. Acta - Mol. Basis
Dis., 2003, 1638, 1-19.
221. J. S. Culp and L. G. Butler, Arch. Biochem. Biophys., 1986, 246, 245-249.
222. R. Gijsbers, H. Ceulemans, W. Stalmans and M. Bollen, J. Biol. Chem., 2001,
276, 1361-1368.
223. E. Bischoff, T. Tran-Thi and K. F. A. Decker, Eur. J. Biochem., 1975, 51, 353-
361.
224. L. H. Schliselfeld, J. van Eys and O. Touster, J. Biol. Chem., 1965, 240, 811-
818.
225. W. Hensel, D. Rakow and W. Christ, Anal. Biochem., 1975, 68, 128-137.
References
259
226. D. W. Kahn and B. M. Anderson, J. Biol. Chem., 1986, 261, 6016-6025.
227. D. J. Wise, C. D. Anderson and B. M. Anderson, Vet. Microbiol., 1997, 58, 261-
276.
228. R. K. Haroz, J. S. Twu and R. K. Bretthauer, J. Biol. Chem., 1972, 247, 1452-
1457.
229. M. Bartkiewicz, H. Sierakowska and D. Shugar, Eur. J. Biochem., 1984, 143,
419-426.
230. J. Wierzchowski, H. Sierakowska and D. Shugar, Biochim. Biophys. Acta -
Protein Struct. Mol. Enzymol., 1985, 828, 109-115.
231. H.-p. Cheng, S. Wei, L.-p. Wei and A. Verkhratsky, Acta Pharmacol. Sin., 2006,
27, 767-772.
232. H. C. Lee, R. Graeff and T. F. Walseth, Biochimie, 1995, 77, 345-355.
233. H. C. Lee, Physiol. Rev., 1997, 77, 1133-1164.
234. F.-J. Zhang, Q.-M. Gu and C. J. Sih, Bioorg. Med. Chem., 1999, 7, 653-664.
235. Q.-M. Gu and C. J. Sih, J. Am. Chem. Soc., 1994, 116, 7481-7486.
236. H. C. Lee and R. Aarhus, Cell Regul., 1991, 2, 203-209.
237. H. C. Lee and R. Aarhus, Biochim. Biophys. Acta - Protein Struct. Mol.
Enzymol., 1993, 1164, 68-74.
238. C. Cakir-Kiefer, H. Muller-Steffner and F. Schuber, Biochem. J., 2000, 349,
203-210.
239. R. M. Graeff, T. F. Walseth, K. Fryxell, W. D. Branton and H. C. Lee, J. Biol.
Chem., 1994, 269, 30260-30267.
240. H. Kim, E. L. Jacobson and M. K. Jacobson, Mol. Cell. Biochem., 1994, 138,
237-243.
241. H. l. n. M. Muller-Steffner, A. l. Augustin and F. Schuber, J. Biol. Chem., 1996,
271, 23967-23972.
242. M. Ziegler, D. Jorcke, J. Zhang, R. Schneider, H. Klocker, B. Auer and M.
Schweiger, Biochemistry, 1996, 35, 5207-5212.
243. P. H. Pekala and B. M. Anderson, J. Biol. Chem., 1978, 253, 7453-7459.
244. H. M. Muller, C. D. Muller and F. Schuber, Biochem. J., 1983, 212, 459–464.
245. H. C. Lee, R. M. Graeff, C. B. Munshi, T. F. Walseth and R. Aarhus, in Methods
Enzymol., ed. J. W. S. C. W. Donald B. McCormick, Academic Press, 1997, pp.
331-340.
References
260
246. S. Zahringer, Novel non-natural dinucleotides as chemical tools for biological
studies, 2012, Diploma thesis.
247. S. P. Jackson and J. Bartek, Nature, 2009, 461, 1071-1078.
248. N. Dwivedi, D. Dube, J. Pandey, B. Singh, V. Kukshal, R. Ramachandran and
R. P. Tripathi, Med. Res. Rev., 2008, 28, 545-568.
249. J. M. Pascal, Curr. Opin. Struct. Biol., 2008, 18, 96-105.
250. A. Wilkinson, H. Sayer, D. Bullard, A. Smith, J. Day, T. Kieser and R. Bowater,
Proteins, 2003, 51, 321-326.
251. J. Y. Lee, C. Chang, H. K. Song, J. Moon, J. K. Yang, H.-K. Kim, S.-T. Kwon
and S. W. Suh, EMBO J., 2000, 19, 1119-1129.
252. L. K. Wang, H. Zhu and S. Shuman, J. Biol. Chem., 2009, 284, 8486-8494.
253. K. S. Gajiwala and C. Pinko, Structure, 2004, 12, 1449-1459.
254. H. Zhu and S. Shuman, J. Biol. Chem., 2005, 280, 12137-12144.
255. J. Nandakumar, P. A. Nair and S. Shuman, Mol. Cell, 2007, 26, 257-271.
256. L. K. Wang, P. A. Nair and S. Shuman, J. Biol. Chem., 2008, 283, 23343-23352.
257. A. Wilkinson, A. Smith, D. Bullard, M. Lavesa-Curto, H. Sayer, A. Bonner, A.
Hemmings and R. Bowater, Biochim. Biophys. Acta - Proteins Proteomics,
2005, 1749, 113-122.
258. T. I. Meier, D. Yan, R. B. Peery, K. A. McAllister, C. Zook, S.-B. Peng and G.
Zhao, FEBS J., 2008, 275, 5258-5271.
259. R. P. Tripathi, J. Pandey, V. Kukshal, A. Ajay, M. Mishra, D. Dube, D. Chopra,
R. Dwivedi, V. Chaturvedi and R. Ramachandran, MedChemComm, 2011, 2,
378-384.
260. G. Ciarrocchi, D. G. MacPhee, L. W. Deady and L. Tilley, Antimicrob. Agents
Chemother., 1999, 43, 2766-2772.
261. W. Gu, T. Wang, F. Maltais, B. Ledford, J. Kennedy, Y. Wei, C. H. Gross, J.
Parsons, L. Duncan, S. J. Ryan Arends, C. Moody, E. Perola, J. Green and P. S.
Charifson, Bioorg. Med. Chem. Lett., 2012, 22, 3693-3698.
262. T. Wang, L. Duncan, W. Gu, H. O‟Dowd, Y. Wei, E. Perola, J. Parsons, C. H.
Gross, C. S. Moody, S. J. Ryan Arends and P. S. Charifson, Bioorg. Med. Chem.
Lett., 2012, 22, 3699-3703.
263. L. Miesel, C. Kravec, A.-T. Xin, P. McMonagle, S. Ma, J. Pichardo, B. Feld, E.
Barrabee and R. Palermo, Anal. Biochem., 2007, 366, 9-17.
264. A. Bondi, J. Phys. Chem. B, 1964, 68, 441-451.
References
261
265. D. L. Miles and H. Eyring, Int. J. Quantum Chem., 1978, 14, 173-190.
266. T. Kouzarides, EMBO J., 2000, 19, 1176-1179.
267. J. Schemies, U. Uciechowska, W. Sippl and M. Jung, Med. Res. Rev., 2010, 30,
861-889.
268. T. Finkel, C.-X. Deng and R. Mostoslavsky, Nature, 2009, 460, 587-591.
269. B. J. North and E. Verdin, Genome Biol., 2004, 5, 224.
270. N. Dali-Youcef, M. Lagouge, S. Froelich, C. Koehl, K. Schoonjans and J.
Auwerx, Ann. Med., 2007, 39, 335-345.
271. K. T. Howitz, K. J. Bitterman, H. Y. Cohen, D. W. Lamming, S. Lavu, J. G.
Wood, R. E. Zipkin, P. Chung, A. Kisielewski, L.-L. Zhang, B. Scherer and D.
A. Sinclair, Nature, 2003, 425, 191-196.
272. Y. Ye and A. Godzik, Nucleic Acids Res., 2004, 32, 582-585.
273. Y. Ye and A. Godzik, Protein Sci., 2004, 13, 1841-1850.
274. B. J. North, B. Schwer, N. Ahuja, B. Marshall and E. Verdin, Methods, 2005,
36, 338-345.
275. M. Magrane and U. Consortium, Database, 2011, 2011.